characterisation of antigen-presenting cells in the murine...

Characterisation of antigen-presenting

cells in the murine female reproductive

tract and its draining lymph nodes

Rebecca Ellen Roche

MSc by Research

The University of York

Biology

June 2011

Characterisation of antigen-presenting cells in the murine female reproductive tract

and its draining lymph nodes

i

Abstract

Despite the global burden of sexually transmitted diseases, the immunology of the

female reproductive tract is poorly understood. An understanding of how the

distribution of antigen presenting cells in the tissue and cross talk between cell types

both at the site of antigen uptake and in the draining lymph nodes will be important

for the design of new tools to manipulate local immune responses. The murine

estrous cycle is characterised by large changes in the architecture of the vaginal and

cervical epithelia in response to hormonal changes. Here, I show that the

distribution of APCs in the epithelia of the vagina and cervix is not uniform.

Neutrophils infiltrate the tissue in large numbers during the progesterone high stages

of the murine estrous cycle, but this does not affect DC localisation. The iliac lymph

nodes that drain the reproductive tract are structurally similar to other mucosal

lymph nodes, but DCs are reduced in frequency compared to LNs draining other

sites. RALDH expression, a marker for retinoic acid metabolism was also reduced

in iliac compared to other peripheral lymph nodes. The murine lower female

reproductive tract and its draining lymph nodes are, therefore, distinct from other

mucosal tissues and lymph nodes and warrant further investigation.



ii

Table of Contents

Abstract .........................................................................................................................i

Table of Contents .........................................................................................................ii

List of Tables.............................................................................................................. iii

List of Figures .............................................................................................................iv

Acknowledgements ......................................................................................................v

Author’s Declaration...................................................................................................vi

1. Introduction...........................................................................................................1

1.1. Defence of mucosal tissues ...............................................................................1

1.2. Initiation of the immune response .....................................................................2

1.3. Effects of the local tissue environment on APCs ..............................................3

1.4. Characteristics of mucosal immune responses ..................................................4

1.5. Challenges of studying immunology in the FRT ..............................................5

1.6. Aims ..................................................................................................................6

2. Materials and Methods.............................................................................................8

2.1. Mice...................................................................................................................8

2.2. Depo-Provera treatment and vaginal smears.....................................................8

2.3. Serial sectioning of the FRT..............................................................................8

2.4. Haematoxylin and eosin staining.......................................................................8

2.5. Immunohistochemistry......................................................................................9

2.6. Quantification of immunohistochemically stained tissue sections..................10

2.7. Stereo imaging of CD19CreR26REYFPxC57BL/6xC57BL/6 mice ..............10

2.8. Isolation of lymph node cells and surface staining for flow cytometry ..........11

2.9. ALDEFLOUR staining....................................................................................11

2.10. Statistical Analysis ........................................................................................12

3. Histological characterisation of the lower FRT over the murine estrous cycle .....13



iii

3.1. Introduction .....................................................................................................13

3.2. Results .............................................................................................................15

3.3. Discussion .......................................................................................................18

4. Localisation of innate immune cells in the FRT ....................................................28

4.1. Introduction .....................................................................................................28

4.2. Results .............................................................................................................29

4.3. Discussion .......................................................................................................32

5. Characterisation of the lymph nodes draining the reproductive tract ....................42

5.1. Introduction .....................................................................................................42

5.2. Results .............................................................................................................43

5.3. Discussion .......................................................................................................45

6. Discussion ..............................................................................................................51

6.1. General conclusions ........................................................................................51

6.2. Disadvantages of Mouse Models.....................................................................53

6.3. Future Work ....................................................................................................53

Abbreviations .............................................................................................................55

References ..................................................................................................................58

List of Tables

Table 2.1.......................................................................................................................9

Table 2.2.....................................................................................................................11

Table 3.1.....................................................................................................................16



iv

List of Figures

Figure 1.1......................................................................................................................7

Figure 3.1....................................................................................................................20

Figure 3.2....................................................................................................................21

Figure 3.3....................................................................................................................22

Figure 3.4....................................................................................................................23

Figure 3.5....................................................................................................................24

Figure 3.6....................................................................................................................25

Figure 3.7....................................................................................................................26

Figure 3.8....................................................................................................................27

Figure 4.1....................................................................................................................35

Figure 4.2....................................................................................................................36

Figure 4.3....................................................................................................................37

Figure 4.4....................................................................................................................38

Figure 4.5....................................................................................................................39

Figure 4.6....................................................................................................................40

Figure 5.1....................................................................................................................46

Figure 5.2....................................................................................................................47

Figure 5.3....................................................................................................................48

Figure 5.4....................................................................................................................49



v

Acknowledgements

First I must express my heart-felt gratitude to everyone in the CII and for their

friendship, kindness and support over the past years.

I would also like to thank Julie Knox and the members of the Biology Graduate

Board for giving me the chance to achieve this.

I am forever grateful to my supervisors Paul Kaye and Marika Kullberg for their

invaluable help, support and patience, as well as my thesis advisory panel, Charles

Lacey and Henry Leese, for all their feedback and suggestions.

I am particularly indebted to Katrein Schäfer for help with vaginal smears and

histology and for generously providing supplementary data for this report. Special

thanks must also go to Najmeeyah Brown for help with mouse work, Jane Dalton,

Lynette Beattie and Paul Mitchell for help with immunohistochemistry and flow

cytometry, Priyanka Narang for help with lymph nodes and Roger Leigh for always

knowing a better software package.

Most of all; eternal thanks to Alex Morris, without whom I would never have got

this far.



vi

Author’s Declaration

I declare that this is my own, original work with the exception for Figure 3.2A,

Figure 3.3A and B, Figure 3.6A in which staining and imaging were done by Katrein

Schäfer as well as all work for Figure 4.5.



1

1. Introduction

Sexually transmitted infections (STIs) are a worldwide health problem [1, 2]. In

1999 there were 340 million new cases of STIs worldwide [3]. This equates to a

huge burden of disease. They cause high mortality; AIDS related deaths account for

1.8 million deaths globally per year [4]. STIs can also cause complications such as

infertility and spontaneous abortion and increase risks of developing other diseases;

for example 95% of cervical cancers are associated with previous HPV infection [5].

STIs are a growing problem with rates of new infections going up in the UK [6].

Protective immune responses against STIs may rely on long-lived, effective local

responses in the female reproductive tract (FRT) to prevent infection. However,

these have proven difficult to manipulate, as both systemic and intravaginal

immunization strategies are unreliable at initiating protective immune responses in

the FRT [7, 8]. Despite the need for comprehensive research, immunity in the

female reproductive tract (FRT) remains under-studied.

1.1. Defence of mucosal tissues

The mucosal surfaces of the gastro-intestinal, respiratory and genito-urinary tracts

are the main sites of pathogen entry. They must act as barriers to potential pathogens

[9], but are also populated by harmless, and in some cases beneficial, commensal

organisms [7, 9-11]. Immunity in mucosal tissues must balance prevention of

disease with tempered responses to the vast majority of microorganisms without

disrupting the normal functioning of the mucosal tissues [12]. In mucosal tissues the

majority of antigen elicit tolerogenic responses, which are characterised by

regulatory T cells (Tregs) and anti-inflammatory cytokines such as IL-10 and TGF-β

[13].

Mucosal tissues facilitate the acquisition of essential nutrients, water and oxygen

from the environment and, therefore, must balance the need to allow passage of

some molecules while preventing the invasion of potential pathogens. To limit

pathogen invasion, mucosal secretions contain a variety of antimicrobial agents. For

example there are antimicrobial peptides in saliva, urine, intestinal fluid [14],



2

seminal plasma [15] and cervicovaginal fluid [2, 9, 12, 14]. The stomach produces

acid and the cervicovaginal fluid is also slightly acidic [2, 16]. Mucosal tissues

produce a great deal of mucus [2, 12], which captures organisms. Mechanical

activity, such as the mucociliary escalator in the lung [17], helps to eliminate many

pathogens.

There are many commensal organisms living on mucosal tissues. Some are

beneficial, as in the microbiota of the gastro-intestinal tract [18, 19] and Lactobacilli

[20, 21] in the genitourinary tracts [2]. Commensals may compete with pathogens

for attachment to mucosal surfaces, restricting their colonisation, however the

normal flora is not always beneficial. Candida species, which are normally non-

invasive commensals can cause invasive infections in response to changes in the

vaginal environment brought on by hormonal changes [22].

1.2. Initiation of the immune response

Cells of the immune system can be divided into innate and adaptive immune cells.

Innate phagocytic cells such as macrophages and neutrophils, which engulf and

destroy pathogens, recognise broad categories of pathogen and respond rapidly to

infection. Adaptive cells, such as T and B cells recognise, species or strain specific

protein antigens and tailor the immune response to that particular organism. They

mediate more efficient responses both qualitatively and quantitatively to such an

extent that they can confer life-long systemic protection from re-infection. There are

large numbers of T and B cells, each with different antigen specificities.

At mucosal sites antigen-presenting cells (APCs) survey the local environment for

signs of infection and tissue damage. Many cell types can be APCs including

macrophages, B cells and neutrophils [23-25]. Dendritic cells (DCs) are phagocytic

cells that bridge the gap between the innate and adaptive immune systems. They are

‘professional’ antigen presenting cells that reside within tissues and lymph nodes.

Their primary function is to acquire ‘foreign’ peptide antigen and initiate the

adaptive immune response. They express an array of pathogen recognition receptors

(PRRs), such as Toll-like receptors (TLRs) [26, 27], that have a broad specificity for

highly conserved pathogen associated molecular patterns (PAMPs) [2]. Acquisition



3

of antigen and ligation of PRRs results in DC maturation and migration to draining

lymph nodes (dLN) where they present antigen to naïve T cells.

T cells recognise a specific peptide antigens bound to major histocompatibility

complex (MHC) on APCs. APCs express co-stimulatory molecules, such as CD40

which lower the threshold for activation. They also direct the homing of T cells to

different tissues. For example APCs induce T cell homing to mucosal tissues by

production of retinoic acid, which induces the upregulation of mucosal addressins

(discussed further in chapter 5). Finally, APCs produce cytokines that direct T cell

differentiation into functionally different subtypes of effector T cells, which mediate

different types of immune response. After clearance of infection some antigen-

specific memory T cells patrol the mucosal sites and LNs ready to respond to re-

infection [28].

1.3. Effects of the local tissue environment on APCs

There is increasing appreciation within the field of immunology of the role played by

non-haematopoietic cells in all stages of immunity [29-39]. DC maturation is

affected by the local environment. Cytokines, stress molecules and cellular

interactions can all affect DCs [27]. Epithelial cells of the oral mucosa [33], gut [11,

19], FRT [40, 41] and lungs [32, 42], as well as epidermal and endothelial cells [39]

also express PRRs and on ligation of those PRRs they can produce chemokines,

which recruit APCs and other immune cells, and cytokines, that can influence the

maturation of DCs [11, 32, 42]. APCs are also sensitive to molecular markers

produced by damaged epithelial cells, which affect maturation and cell recruitment.

These can be cytokines, chemoattractants, alarmins such as HMGB1 or heat shock

protein which are produced by infected or damaged cells, or intracellular molecules,

which are only found outside cells after necrotic cell death [31, 43]. There are

different stress indicator molecules produced after different severities of epithelial

damage, which bias the immune response in different ways; promoting a regulatory

response after mild damage and an inflammatory response after severe damage [31,

32]. Epithelial cells in the intestine condition DCs and macrophages towards

tolerogenic responses to commensal bacteria in order to maintain gut homeostasis

[11].



4

It is important not to view immune responses in terms of interactions among innate

and adaptive immune cells and pathogens on a background of tissue stromal cells,

but to think of the functioning of the organ and its resident stromal cells and how it

recognises pathogens, recruits and directs immune cells and resolves infection while

retaining its normal functioning. If the small intestine is infected the main event is

not the immune response to the infection, but the continuing absorption of food. It is

therefore important to understand the immune response in the context of the tissue it

occurs within.

1.4. Characteristics of mucosal immune responses

The majority of antigen that enters the body through mucosal surfaces induces a

tolerance response. This has been best characterised by the immunosuppressive and

anergic responses to orally administered antigen - so called oral tolerance [44]. The

antibody response at mucosal surfaces is usually dominated by IgA [13] with lower

levels of IgG and IgE [7]. The predominant antibody in serum is IgG [8]. In the

small intestine B cells switch to IgA production in the Peyer’s patches (PPs) under

the influence of TGF-β and IL-10 [13].

T cells primed in the small intestine express distinct mucosal addressins and

chemokine receptors, which allow them to traffic to the intestinal mucosa. T cells

from the MLNs upregulate α4β7 integrin, which binds MAdCAM-1 expressed on

mucosal vasculature. T and B cells express CCR9, which binds CCL25 expressed

by intestinal epithelial cells [7, 13].

In PPs and MLNs there are distinct phenotypes of DCs. There are unusually large

numbers of CD8α-CD11b- DCs and lower percentages of CD8α-CD11b+ DCs

compared to the spleen where the majority of DCs are CD8α-CD11b+ [13]. CD8α-

CD11b+ DCs in the PPs preferentially produce IL-10 compared to splenic CD8α-

CD11b+ DCs, which produce IL-12 under the same conditions. DCs in the MLNs

produce IL-10 and TGF-β in response to oral antigen and promote upregulation of

α4β7 on T cells [13].



5

1.5. Challenges of studying immunology in the FRT

The tissue of the lower FRT consists of layers of squamous epithelial cells, similar to

skin, with connective tissue underneath (Fig 1.1) [28, 45]. The epithelia are covered

in mucus. In humans the epithelia is non-keratinized, but in mice the outer layers

become keratinized [45, 46].

The FRT is a unique mucosal environment. The FRT undergoes homeostatic

remodelling in response to sex hormones in mammalian reproductive cycles, which

results in dramatic cyclic changes in the epithelia and lamina propria of the uterus,

cervix and vagina. The FRT must prevent infection while at the same time

promoting conception and gestation of the genotypically different foetus. This

presents a challenge for the immune system. The issue of balancing the

reproductive function against protection from disease is particularly problematic

after ovulation. The FRT must become permissive to sperm to increase the chance

of fertilization, but sexual intercourse is also when the female is at highest risk of

infection by sexually transmitted pathogens. The environment of the vagina changes

to allow insemination. It becomes less acidic, the mucus becomes less thick and

more permissive [47].

Unlike other mucosal sites, the FRT tissue and the tissue resident immune cells are

similar to those of the skin. There are Langerhans’ cells (LCs) in the epithelia with a

separate DC population in the lamina propria. As has been shown in the skin [48,

49], different DC subtypes have different roles in immunity [50]. Submucosal DCs

rather than the LCs, migrate to the draining lymph nodes and present antigen to T

cells [50]. The humoural immune response in the FRT is different to other mucosal

surfaces. Like the gut, the FRT has IgA in secretions; however unlike the gut there

are much higher levels of IgG in cervicovaginal secretions [8, 40, 51-54]. IgG in

cervicovaginal secretions is produced locally by antigen secreting cells and comes

from the blood [8, 40, 51-54].

There is some evidence that hormones and their effects on stromal and epithelial

cells in the FRT can affect immune responses [40, 55-60] through changes in PRR

expression [61], cell recruitment, antibody levels, antigen presentation[40, 55, 58,



6

59, 62], T cell responses and susceptibility to STDs [58-60] over the estrous cycle or

after hormone treatment (discussed further in chapter 4).

Like in the intestinal mucosa, the FRT can induce tolerance [63], but it appears to be

estrous cycle stage specific. There was no difference in antigen-trafficking to the

draining lymph nodes or in the serum antibody response at different cycle stages, but

spleen and lymph node cells also showed significantly less proliferation in response

to restimulation after priming in estrus [63].

1.6. Aims

Relative to other mucosal sites little is known about the immunology of the genital

tract. The changeable conditions in the FRT pose a challenge to experimental

design. Rats & mice have different reproductive cycles to humans and therefore

questions remain about their suitability for translational research.

The aim of this study was to characterise the murine FRT in steady state to identify

some of the factors which affect the initiation of the immune response to antigen.

Specifically;

i. To use immunohistochemistry to characterise the FRT at different stages of

the cycle.

ii. To examine the distribution of antigen presenting cells in the vagina and

cervix

iii. To compare the FRT dLNs (iliac LNs) to other mucosal/non-mucosal LNs to

try to identify factors, which may affect antigen presentation and the

downstream adaptive immune response.



7

Figure 1.1: The human female reproductive tract.

The FRT can be split into two: The lower FRT, consisting of the vagina and

ectocervix and the upper FRT consisting of the endocervix, uterus and ovaries. The

lower FRT is covered in layers of squamous epithelial cells, while the upper FRT is

covered in a simple columnar epithelial layer. Figure from [45].



8

2. Materials and Methods

2.1. Mice

6-10 week old female C57BL/6 (B6) or B6.CD45.1 mice (Charles River UK) were

housed under specific-pathogen free conditions Mice were kept in a continuous 12

hour light/dark cycle and were caged only with females from weaning. 8 week old

female CD19CreR26REYFPVaDsRedxC57xC57 mice were bred under specific-

pathogen free conditions. All procedures were done in compliance with the Animal

(Scientific Procedures) Act 1986.

2.2. Depo-Provera treatment and vaginal smears

Mice were injected subcutaneously in the base of the tail with 100 µl of 30 mg/ml

Depo-ProveraTM (Pharmacia) in sterile saline (Baxters, UK) 5 days prior to use.

Vaginal smears were taken using a pipette and approximately 30-50 µl sterile saline

then placed on polylysine slides (Fisher) with or without further dilution, depending

on viscosity of the mucus. Slides were air dried then haematoxylin and eosin (H&E)

stained.

2.3. Serial sectioning of the FRT

Reproductive tracts from mice were divided into lower vagina, upper vagina, and

cervix and snap frozen in OCT embedding medium (TissueTek) in plastic

Cryomolds (TissueTek) and stored at -80°C. Transverse 7 μm serial sections from

the lower vagina, upper vagina and cervix were cut on a cryostat and placed on

polylysine slides (Fisher). Approximately 6 slides (approx. 48 tissue sections) were

prepared sequentially. 60-100 µm of tissue was removed before another 6 slides

were prepared to allow representative sampling of the whole lower FRT. Slides

were air-dried then stored at -20°C ready for H&E staining or

immunohistochemistry.

2.4. Haematoxylin and eosin staining

Air-dried slides were stained in Harris haematoxylin (Sigma) for 5 minutes and then

washed in cold running water for 5 minutes. Slides were dipped 12 times in 0.5%



9

eosin then dipped in distilled water. Slides were then dipped 10 times in 50%

ethanol, 10 times in 75% ethanol, left in 95% ethanol for 30 seconds followed by 1

minute in 100% ethanol. Slides were air dried then mounted in DePex (VWR BHD

Prolabo) with a cover slip. Images were acquired on Zeiss Axioplan microscope

using an Optronics camera and brightness and contrast were adjusted using

Photoshop, as required.

2.5. Immunohistochemistry

Sections were fixed in ice cold acetone for 5 minutes, washed in PBS 0.05% BSA

(Fisher) and then blocked in PBS 0.05% BSA containing 5% Rat (Sigma), or

Hamster (MP Biomedicals) serum or blocking IgG for at least 1 hour. An avidin-

biotin blocking kit (Invitrogen) was used according to the instructions, when

biotinylated antibodies were used. Briefly, sections were blocked with β-biotin (10

min), washed (PBS 0.05%BSA) and then incubated with avidin (10 minutes)

followed by 3 washes. Slides were then stained with CD11c, CD68 or Gr-1

antibodies (see table) or appropriate isotype controls in blocking buffer for 45

minutes. Slides were washed as above. For experiments using biotinylated

antibodies Streptavidin-AlexaFluor546 or Streptavidin-AlexaFluor488 (both

eBioscience) diluted 1:300 in blocking buffer was then added and incubated at room

temperature for 30 minutes. The slides were then washed 3 times as above and then

once in PBS. Slides were counterstained with 1 mg/ml DAPI, for 5 minutes, if

required, washed in PBS twice and then mounted in Prolong Gold anti-fade reagent

(Fisher) with a cover slip. Images were captured using a Zeiss inverted 510 Meta on

Axiovert 200M confocal microscope and analysed using Zeiss LSM Image Browser

software and cells were counted using ImageJ software. Stitched images of LNs

were made using Fiji software.

Antibody Fluorochrome Isotype control Dilution Supplier

CD11c Alexafluor 647 Hamster IgG 1:200 eBioscience

CD68 Alexafluor 647 Rat IgG2a 1:100 eBioscience

MHCII biotinylated Rat IgG2b 1:200 eBioscience

MHCII Alexafluor 450 Rat IgG2b 1:200 eBioscience

GR-1 biotinylated Rat IgG2b 1:200 eBioscience

gp38 Alexafluor 488 Hamster IgG 1:200 eBioscience



10

CD31 biotinylated Rat IgG2a 1:200 eBioscience

Table 2.1 Antibodies used for immunohistochemistry

2.6. Quantification of immunohistochemically stained tissue

sections

2.6.1. Pixel counts

For each field of view lines were drawn around the epithelia using Photoshop. The

epithelia were selected and the total pixel counts performed. The green, red and

yellow channels were selected in turn and pixel counts performed for each. The

pixel count for one channel was divided by the total pixels (x100) to give a

percentage. For the total CD11c or MHCII cells the red and green channels

respectively were added to the yellow channel and expressed as a percentage of total

pixels.

2.6.2. Cell counts

The lengths of the basal layers of the epithelia were measured using LSM software.

The number of cells in the epithelia were counted using ImageJ software and the

results were expressed as cells/mm.

2.6.3. Determining neutrophil density

Blind analysis was done using LSM software. Images showing only GR-1+ and

DAPI staining were designated as high or low neutrophil areas. Greater than 50% of

the outer epithelial length positive for Gr-1 was designated as neutrophil high.

Images showing only CD11c+ and DAPI staining channels were used for DC counts.

2.7. Stereo imaging of CD19CreR26REYFPxC57BL/6xC57BL/6

mice

Single images of mice were taken under white, red fluorescent and green fluorescent

light in turn on a stereo microscope (Zeiss) and captured using a CCD camera.

Individual images were overlaid in ImageJ.



11

2.8. Isolation of lymph node cells and surface staining for flow

cytometry

Lymph nodes from multiple mice were pooled into groups of cervical lymph nodes,

iliac lymph nodes, mesenteric lymph nodes and a group of ‘other’ lymph nodes,

which included axillary and inguinal lymph nodes. Pooled LNs were diced into

small pieces with a scalpel and forceps then washed in PBS. These were incubated

for 25 minutes at 37°C in 1 ml of enzyme mix containing 1.8 Wünsch units/ml

Liberase TL (Roche) and 0.5 mg/ml DNaseI (Roche) in PBS. The resulting digests

were then passed through a 70 µm cell strainer (BD Biosciences). Cells were

washed (1300 rpm for 5 minutes) twice in PBS 1% FCS. Viable cell counts were

determined by Trypan blue exclusion using a haemocytometer. Pooled cells from

lymph nodes were blocked in Fc block (eBioscience 0.5 µg/ml) in PBS 1% FCS for

15 minutes on ice. Cells were washed in PBS 1% FCS as above then split into 2x106

per sample. Cells were stained for CD11c, MHCII, CD11b, CD103 and CD45.2 or

with appropriate isotype controls (See table 2.2) and incubated on ice, covered in foil

to protect from light for 25 minutes. Cells were washed twice in PBS 1% FCS as

above. Cells were fixed in 4% PFA for 20 mins on ice and were stored at 4°C until

analysis. Samples were analysed on a CyAn flow cytometer using Summit software

(Beckman Coulter).

Antibody Fluorochrome Isotype

control

Dilution Supplier

CD11c PE-Cy7 Hamster IgG 1:400 eBioscience

MHCII e450 Rat IgG2b 1:400 eBioscience

MHCII APC-Cy7 Rat IgG2b 1:400 eBioscience

CD45.2 APC-780 Rat IgG2a 1:200 eBioscience

CD11b e450 Rat I gG2b 1:200 eBioscience

CD103 PE Rat IgG2a 1:200 eBioscience

Table 2.2 Antibodies used for flow cytometry staining

2.9. ALDEFLOUR staining

To identify cells containing active RALDH enzyme activity, ALDEFLUOR reagents

(Stemcell Technologies) were used. Cells were stained for surface markers as above

but without fixation. After surface staining, cells were washed in PBS 1% FCS as



12

above then resuspended in 50 µl ALDEFLUOR assay buffer. 50 µl ALDEFLUOR

substrate or DEAB inhibitor diluted 1:15 in ALDEFLUOR substrate were added to

appropriate wells and incubated in foil at 37°C 5% CO2 for 30 minutes. Samples

were washed in ALDEFLUOR buffer then resuspended in PBS. 1 µl of

LIVE/DEAD Fixable aqua dead cell stain (Invitrogen) was added to each sample

then incubated on ice for 30 minutes in foil. Samples were washed in PBS then

resuspended in PBS 1% FCS before analysis on a CyAn flow cytometer using

Summit software (Beckman Coulter).

2.10. Statistical Analysis

Statistical analysis was performed using GraphPad InStat 3 software. Nonparametric

Mann-Witney-tests were used for all analysis.



13

3. Histological characterisation of the lower FRT over the

murine estrous cycle

3.1. Introduction

Mice are the common model for immunology and infection research because they

are a relatively cheap and easy to use mammalian species that are genetically similar

to humans and other mammals. They also have the advantage of the huge range of

genetically defined or altered mouse strains available. These allow researchers to

design experiments that assess the role of specific genes, which could not be done in

humans. Research in mouse models is often used to draw conclusions about human

immunology, so it is important to have a thorough understanding of the similarities

and differences between them and what this could mean for the application of mouse

research to human research. Importantly, humans and mice have different

reproductive biology. Human females of reproductive age, not receiving hormonal

contraception have a menstrual cycle which lasts around 28 days. It is characterised

by a period of endometrial growth, known as the follicular phase, followed by

ovulation. This is followed by the luteal or secretory phase, where the ovaries

produce large amounts of hormones that facilitate implantation and early growth of

the fertilised ovum. If fertilisation does not occur the ovum degenerates and

menstruation occurs, in which the endometrium is sloughed off and is expelled from

the body [16]. Conversely, mice and rats have estrous cycles. The main difference

between mammalian menstrual and estrous cycles is that in the menstrual cycle the

epithelial cells shed into the lumen are discharged out of the vagina, whereas in the

estrous cycle the epithelial cells are broken down and reabsorbed [16].

Mice have non-seasonal polyestrous cycles and spontaneously ovulate [16, 64],

which means that, similar to humans, they ovulate multiple times throughout the

year at regular intervals and do not require sexual stimulation to ovulate. Unlike

many other estrous animals such as sheep and dogs, the murine estrous cycle is not

driven by seasonal cues and mice will not naturally enter a anestrus stage in which

the animal is unable to reproduce [16].



14

The murine estrous cycle lasts 4 to 5 days [64] and ovulation occurs spontaneously

whether mating has occurred or not [64]. It comprises a period of preparing the

tissue for nourishing a fertilised egg, before a period of ‘heat’ where the animal is

receptive to mating and ovulation occurs. If fertilisation does not occur the

reproductive tract undergoes a period of shedding before restarting the cycle. The

estrous cycle can be split into 4 stages; proestrus in which the ovarian follicles grow

and the uterine epithelium thickens, estrus in which ovulation occurs, metestrus

begins when the uterine epithelium sheds and diestrus when the epithelia is at its

thinnest before the cycle restarts [64, 65]. Metestrus and diestrus are sometimes

referred to as metestrus-1 and -2 or diestrus-1 and -2 [64], however, in this study the

metestrus will be used to refer to the stage immediately after estrous followed by

diestrus. Proestrus and estrus are equivalent to the follicular phase of the human

menstrual cycle and metestrus and diestrus are similar to the luteal phase.

Events in the ovaries are almost identical in the estrous and menstrual cycles and the

kinetics of luteinising hormone (LH) and follicle-stimulating hormone (FSH) levels

in the blood stream follow almost identical patterns [16, 64, 66]. FSH produced in

the pituitary promotes maturation of the ovarian follicle. LH causes the maturing

follicle to produce estrogen, which results in increased LH production through a

positive feedback loop. Rising LH causes ovulation [64]. The ovarian follicle then

becomes a corpus luteum and secretes hormones that promote survival and growth of

the fertilised ovum. The corpus luteum degenerates and the decreasing hormone

concentrations causes epithelial shedding in the uterus [16, 64].

The vaginal epithelium consists of stratified squamous epithelium [2, 16, 41], similar

to skin. The stromal and epithelial cells are under hormonal control. The epithelial

architecture of the uterus, cervix and vagina changes in response to estrogen and

progesterone (Fig3.1). Increased estrogen leads to thickening of the epithelia and

ovulation, while high progesterone is associated with thinning of the epithelia.

Much of the hormone-induced changes in the vaginal tissue are mediated by the

stromal cells in the lamina propria, which control epithelial cell behaviour and

immune cells.



15

3.2. Results

The histology of the FRT over the estrous cycle has been well characterised in rats

[67], but less so in mice. In order to assess the physiological changes in the FRT 6-8

week old female mice were vaginally smeared and these smears were H&E stained

and used to classify each mouse into an estrous cycle stage. C57BL/6 CD45.2 or

C57BL/6 CD45.1 mice were used because they are common laboratory mouse

strains. They differ in the allelic form of CD45, which results in no functional

differences. The two strains are, otherwise, assumed to be identical. The FRTs

were removed and snap frozen in order to preserve the tissue architecture. To

characterise the tissue throughout the tract serial transverse tissue sections of the

vagina and cervix were H&E stained (Fig 3.2).

3.2.1. Vaginal Smears

The histology of vaginal smears can be used as a predictor of estrous cycle stage, as

has been shown in rats [68]. To establish a protocol based on the murine estrous

cycle individual mice were smeared everyday for 4 days+ at approximately the same

time every day. H&E stained vaginal smears were assigned into one of the 4 estrous

cycle stages based on guidelines for the reading of rat vaginal smears [68]. Smears

were also taken from high-dose progesterone (Depo-Provera) treated mice at 5 days

post treatment and on some occasions at 6 – 30 days post treatment. The vaginal

smear classification scheme devised is summarised in table 3.1.

In proestrus (Fig 3.3A) the smears contained few cells and were made up almost

exclusively of epithelial cells. The smears contained differing ratios of small,

rounded, nucleated epithelial cells with pale cytoplasm and larger, cornified cells,

which appear pinker, probably due to keratinisation and increased glycogen.

At estrus there were huge numbers of large non-nucleated cornified epithelial cells in

the smear (Fig3.3B). The cornified cells were large, flattened, highly eosinophilic

and clumped together. Other cells were almost completely absent in the smear.

At metestrus the predominant cells in the wash were leukocytes (Fig3.3C). The

majority of leukocytes were polymorphonuclear neutrophils, which can be identified

by their multi-lobed nuclei (open arrows). The smears were generally thick with



16

mucus, causing leukocytes and non-nucleated cornified epithelial cells to clump

together. In some smears the epithelial cells are nearly all large anucleated cornified

cells, but in others more nucleated cells were present in the smear.

At diestrus the majority of cells in the smear were leukocytes with some cornified

and/or nucleated epithelial cells (Fig3.3D). Leukocytes were less clumped in

diestrus compared to metestrus smears, though there was considerable variation.

After Depo-Provera treatment the smears were similar to the smears from

metestrus/diestrus. The majority of cells were leukocytes with smaller numbers of

nucleated epithelial cells (Fig3.3E). There was a large amount of thick mucus

present in the smears which caused clumping of cells; however this did vary from

mouse to mouse.

Table 3.1: Determining estrous cycle stage using vaginal washes.

3.2.2. Histology at proestrus

The vaginal lumen was devoid of cells in the tissue sections, consistent with the

smears which contained few cells (Fig 3.4A). The epithelium was thick at proestrus

and was densely packed with very few visible gaps between epithelial cells. The

epithelium varied in thickness from approximately 5-25 cells thick (Fig 3.4B&C). In

most sections the outermost layer of the epithelium was anucleated, but not

keratinised (Fig 3.4B), but in some there was a highly eosinophilic band in the outer

epithelium showing keratinisation of the outer epithelium into the stratum corneum

(SC) (Fig 3.4C).

Proestrus Estrus Metestrus Diestrus Depo-

Provera

Nucleated epithelial

cells

+ - - + +

Non-nucleated

epithelial cells

- +++ ++ - -

Leukocytes - - +++ ++ ++

Mucus + ++ +++ ++ +++



17

3.2.3. Histology at estrus

Strings of cornified epithelial cell sheets were seen filling the lumen in tissue

sections (Fig 3.5A). Some areas of the epithelium were relatively thick while others

were comparatively thin. In most sections the stratum corneum was partly detached,

but the degree of detachment varied. In some smears leukocytes were seen in the

epithelia and in the submucosa, but almost never in the lumen (Fig 3.5C). There

were very few gaps between epithelial cells. The cervix had relatively less

desquamed epithelial cells in the lumen than the vagina (Fig 3.5D).

3.2.4. Histology at metestrus

The stratum corneum was almost completely detached, but can be seen in the lumen

along with leukocytes in the tissues sections (Fig 3.6A). The outer epithelium

appeared to lose integrity and there was infiltration of leukocytes, which have darker

staining nuclei than the surrounding epithelial cells (Fig 3.6B). Individual

leukocytes were seen in the submucosa and basal layers of the epithelium and

aggregates of cells were observed in the superficial layers of the epithelium,

sometimes forming small foci. Some cells were identified as neutrophils based on

their distinctive multi-lobed nuclei (Fig 3.6C). The outer epithelia layers were

obviously disrupted in the majority of tissue sections and there were gaps between

epithelial cells. Leukocyte infiltration is not uniform throughout the lower

reproductive tract with some areas experiencing rapid infiltration and loss of

epithelial cell layers while other areas have slower infiltration and an extended

period of slow loss of the epithelial layers (Fig 3.6A-C). This makes early and late

metestrus difficult to differentiate. In this study metestrus is characterised as

beginning with the widespread loss of the stratum corneum and infiltration of

leukocytes and ends with extensive loss of deeper epithelial layers, decreasing

leukocytes in the epithelium and formation of the thick, dark outer epithelial layer

characteristic of diestrus (described below and in Fig 3.7A and C).

3.2.5. Histology at diestrus

The lumen was mostly cell free in tissue sections (Fig 3.7B) and the epithelia was

generally thin with only 2-3 cell layer present in some areas, though this varied



18

within the FRT and even within a single field of view. The outer epithelium

contained leukocytes. In some places the leukocytes were in clumps in the

epithelium, but in others they were largely absent or only visible in the lumen.

While the epithelia appeared to be less leaky than in metestrus, they were less tightly

packed than in proestrus or estrus as there were visible gaps between the epithelial

cells (Fig 3.7B). Where the epithelium was thinnest the outer epithelium contained a

thicker solid band of cells (Fig 3.7A and C). The dark nuclear staining in the outer

epithelia implies that it was made up of leukocytes. The cervical epithelia were

generally slightly thinner than the vaginal epithelia. The cells contained large

vacuoles and there were large nucleated cells in the lumen (Fig 3.7D and E).

3.2.6. Histology of progesterone treated mice

The histology of the FRT after progesterone treatment was also characterised

because it will be used later in the study. B6.CD45.1 mice were injected

subcutaneously with 100 µl (30 mg/ml) Depo-Provera. 5 days later the FRTs were

removed and H& E stained. In all sections the epithelia were thin and all layers

contained nucleated cells (Fig 3.8A). Large mucus-filled vacuoles were seen in the

superficial layers of the epithelia (Fig 3.8C) and there were gaps in the basal layer of

the epithelium. Leukocytes, characterised by darker staining nuclei, were still

present under the mucified cell layer (Fig 3.8B). The cervical epithelia were

slightly thinner than the vaginal epithelia and there was more mucification of cells

(Fig 3.8C).

3.3. Discussion

During proestrus the epithelia thickens leading up to estrus, where ovulation occurs.

After ovulation the stratum corneum sloughs off in sheets. While the stratum

corneum desquames, the underlying epithelia integrity appear to remain intact as

there are no visible gaps between epithelial cells. At the very end of estrus

leukocytes begin to infiltrate into the tissue and appear in the epithelia. Widespread

leukocyte infiltration marks the end of estrus and the beginning of metestrus [67]. At

metestrus the squamous layers were less organised, probably due to the influx of

leukocytes. At diestrus the epithelia is thin and has a sticky thicker outer layer that

shows the beginning of mucification.



19

High dose progesterone treatment is a common way of synchronising mice into

diestrus [50, 69]. The smears and tissue sections of diestrus and DP mice were

similar. After Depo-Provera treatment the superficial epithelial cells were more

heavily mucified than at diestrus, consistent with pseudopregnancy, which is

characterised by high progesterone levels causing mucification of the epithelium [67,

70].

The changes in architecture do not happen simultaneously throughout the tract.

Some areas can have characteristics of different estrous cycle stages to other areas. I

found no evidence of consistent variation up or across the tracts in the serial sections

of numerous mice; instead it is likely these differences are due to individual

differences in the geography of the vaginal epithelia and luminal fluid flow. This

leads to some, more exposed, areas sloughing faster than others and may lead to

variations hormone concentrations and cell recruitment in some areas.

The variation within the local environments of the tissue cannot be reflected in the

smears, which provide a rough sample of the outer epithelia and lumen of the lower

FRT. Vaginal smears provide an approximate overview of the tissue, but cannot, for

example, readily determine between an early estrus mouse with the SC only just

starting to desquame or a late estrus mouse with almost complete detachment of the

SC and infiltration of leukocytes into the basal epithelia, both of these smears would

consist exclusively of large cornified epithelial cells. Vaginal smears are indicative

of the state of the tissue, but will never provide the more comprehensive information

that serial tissue sectioning can.

Vaginal smears offer a quick and convenient method of determining estrous cycle

stage, however they are not always clear cut; it is possible to get transitional smears

that are difficult to assign to one stage or another. The rapid and substantial changes

in the vaginal epithelia mean that the outer epithelium is in constant flux. The

cellular composition of vaginal smears is very dynamic and while a 4 stage estrous

cycle seems superficially simpler to classify, it is a little crude and conceals a lot of

the subtle differences between the smears.

There is a great deal of variation between individual mice in the overall cellularity

and viscosity of the smears, which is difficult to factor in to a classification scheme.



20

Vaginal smears are therefore a good indicator of estrous cycle stage and conditions

in the tissue, but it cannot be assumed that they are 100% accurate or that they reflect

every corner of the FRT.

Neutrophils are recruited into the vagina by the chemokine MIP-2, an IL-8

homologue [71] after the drop in estradiol after ovulation [72]. Epithelia cells can

produce inflammatory cytokines/chemokines, which recruit granulocytes from the

blood [40, 71, 73]. It is unknown whether neutrophils are responsible for the

disruption of the outer epithelia and the sloughing of the cell layers under the stratum

corneum or if neutrophil infiltration is a side effect of the loss of cells. It is possible

that the loss of the superficial layers of epithelium and invasion by bacteria/fungi

caused by the increased permeability of the epithelium may trigger stress responses

in epithelial cells. The barrier function of uterine epithelial cells has been shown to

be effected by hormonal changes [41] and it possible that changes in the vaginal

epithelial cell barrier may lead to the penetration of lumen contents into the epithelial

layer causing production of chemokines by epithelial cells.

Figure 3.1: Serum hormone concentrations in mice.

The hormones progesterone (blue) and estrogen (orange) fluctuate over the estrous

cycle. Estrogen peaks at estrus, when ovulation occurs, then drops off. Progesterone

increases after ovulation. Based on data from [74].



21

Figure 3.2: Experimental Procedure.

1. Vaginal smears were taken from C57BL/6 or CD45.1 female mice and H&E

stained to determine cycle stage. 2. The FRTs were removed. 3. Sagittal view of

murine reproductive tract. 4. FRTs were split into cervix, upper vagina and lower

vagina (boxes) and snap frozen. 5. 7 µm serial transverse sections of tissue were cut

on a cryostat. 6. Diagram of transverse tissue section showing vaginal lumen in

centre, with epithelium and surrounding connective tissue.



22

Figure 3.3: Vaginal Smears

A) Proestrus vaginal smear; insert shows a nucleated epithelial cell (arrow head) and

an anucleated epithelial cell (arrow) showing early cornification. B) Estrus vaginal

smear; insert shows cornified epithelial cells. C) Metestrus vaginal smear; insert

shows leukocytes (arrow) and cornified epithelial cell (arrow head). D) Diestrus

vaginal smear; insert shows leukocytes (arrow) and nucleated epithelial cells

(arrowhead). E) Vaginal smear from Depo-Provera treated mouse; insert shows

clumped leukocytes.



23

Figure 3.4: Histology of lower FRT at proestrus.

H&E stained vaginal smears and 7 µm transverse sections of lower FRT tissue from

6-8 week old C57BL/6 or B6.CD45.1 female mice. A) x10 image of vagina. A-C)

Vagina at early proestrus. B) Early proestrus C) Late proestrus; insert shows the

outer epithelium. Ep epithelium, L lumen, SC stratum corneum, SM submucosa



24

Figure 3.5: Histology of lower FRT at Estrus

H&E images of vaginal smear and 7 µm transverse sections of vagina and cervix at

estrus. A) Lower vagina with lumen containing epithelial cell sheets. B) Vagina at

early estrus. C) Vagina at late estrus; insert shows leukocytes in epithelium

(arrows). D) x10 image of upper vagina epithelium (right) and cervical epithelium

(left) showing the difference in the level of epithelial sloughing. Red arrows indicate

sloughed cornified epithelial cells. Ep epithelium, L lumen, SC stratum corneum,

SM submucosa.



25

Figure 3.6: Histology of lower FRT at metestrus.

H&E images of 7 µm transverse sections of vagina at metestrus. A) Lower vagina

B) Vagina with leukocyte infiltration (arrows). C) Vaginal epithelium; insert shows a

close up of neutrophils. Arrows show leukocytes. Ep epithelia, L lumen, SM

submucosa.



26

Figure 3.7: Histology of lower FRT at diestrus.

H&E images of 7 µm transverse sections of vagina and cervix at diestrus. A)

Vagina. B) Vagina at early diestrus with leukocyte infiltration (arrows). C) Vagina

at late diestrus; insert shows the epithelium. D) Cervix. E) Cervix showing

leukocytes in the epithelium and lumen (arrows) and nucleated epithelial cells in the

lumen (arrow heads) .Arrows show leukocytes. Ep epithelium, L lumen, SM

submucosa.



27

Figure 3.8: Histology of lower FRT after high dose progesterone treatment.

CD45.1 female mice were treated with 100 µl Depo-Provera. 5 days later vaginal

smears were taken and H&E stained. FRTs were then frozen and 7 µm sections

were cut and stained with H&E. A) Vagina. B) Vagina showing thinned

epithelium. C) Vagina showing mucification of the outer epithelia. D) Cervix

showing mucification. Ep epithelium, L lumen, SM submucosa.



28

4. Localisation of innate immune cells in the FRT

4.1. Introduction

Estrogen receptor expression is found both in lymphoid organs and bone marrow and

on DCs and macrophages [40]. Differential estrogen levels are thought to affect early

DC development by affecting the differentiation of DC precursors in the bone

marrow [58]. Estrogen has different effects on different DC development pathways,

promoting GM-CSF induced DC development pathways and inhibiting Flt-3L

mediated development pathways [58]. These studies show how systemic estrogen

may affect the immune system as a whole; however it does not highlight the

immediate effects of estrogen on cells within tissues. Macrophages, B cells and T

cells can all express progesterone receptors, although it is unclear whether this is

constitutive or induced as the data comes from pregnant women [21], which raises

the possibility that progesterone can have direct effects on immune cells.

The efficiency of intravaginal vaccination is affected by phase of the estrous cycle

[75, 76], a finding that might have significant implications on FRT vaccination

strategies. There is some evidence that susceptibility to infections in the FRT may

change over the estrous/menstrual cycle [59]. For example mice are more

susceptible to HSV-2 infection in metestrus and diestrus [50, 76], but are more

susceptible to Neisseria gonorrhoeae infection at proestrus [59]. Rhesus macaques

are more susceptible to SIV infection during the progesterone high stage of the

menstrual cycle, whereas estrogen is protective [21, 77]. Women show increased

susceptibility to HIV and other STIs while taking progesterone contraceptives [21,

77] and hormones have effects on susceptibility to candidiasis and Chlamydia [78] .

These differences could be due to a variety of factors. It may simply be the result of

epithelia thinning during progesterone high stages which allows easier access for

pathogens [21, 77], or changes in epithelium permeability [79]. Antibody and

antimicrobial peptide concentrations vary of the course of the estrous cycle, as do

chemokine levels [21, 59]. Estrogen inhibits expression of the chemokine MCP-1 by

stromal cells in the uterus [21]. Cyclic changes in the commensal bacteria,

particularly H2O2-producing Lactobacilli [21], that line the FRT [80] may affect



29

susceptibility to viral infections [77]. T cell and B cell populations in the FRT can

vary over the reproductive cycle, as does IgG and IgA production [21, 77].

Activation of T cells in the presence of progesterone inhibits Th1 responses and

promotes Th2 responses by inducing IL-4 production and inhibiting IL-12

production by APCs [21, 77] [59]. Estrogen has varied affects on immunity; it can

be both pro-inflammatory and have inhibitory affects. In rats, estradiol increases

antibody levels in the uterus, but reduces levels in the vagina [77]. Low doses of

estrogen are pro-inflammatory while high doses are anti-inflammatory [21].

In this chapter the distribution of DC and macrophage in the mouse FRT during

cycle will be compared to previously published studies.

4.2. Results

4.2.1. Dendritic cell and macrophage localisation in the FRT in

different hormonal conditions

The location of APCs within the tissue is important because it will affect their ability

to sample antigen from the lumen and to interact with other cells. To investigate

whether APC localization within the tissue is affected by the hormonal changes the

vaginas from estrus (the estrogen high stage), metestrus (the progesterone high stage

characterized by large infiltration of leukocytes) and progesterone treated mice were

stained for CD11c and CD68 (Fig 4.1).

There were greater numbers of CD11c+ cells in the submucosa compared to the

epithelia both at estrus and metestrus (Fig 4.1). At estrus only the basal layers of

epithelial cells were nucleated and stained DAPI+ (Fig 4.1A). The cornified sheets

of cells auto-fluoresced so can be seen adjacent to the lumen. CD11c+ cells were

almost completely absent. Despite the fact that the epithelium had increased in

thickness, CD11c+ cells were only present in the basal layer or immediately

underneath with processes reaching in between the epithelial cells. No CD11c+ cells

were observed reaching to the luminal edge of the epithelia. There were fewer cells

in the submucosa compared to metestrus. At metestrus CD11c+ interdigitating cells

(Fig 4.1C arrows) were present in both the basal and superficial layers of the

epithelium (Fig 4.1C).



30

CD68+ cells were largely absent from the epithelia in both estrus and metestrus (Fig

4.1B and D). They were present throughout the submucosa at similar levels in both

estrus and metestrus. There were no obvious differences in CD68+ localisation

during cycle.

4.2.2. Localisation of CD11c and MHCII expressing cells throughout

the tract

To see if there are any differences in the distribution of APCs in the lower FRT serial

tissue sections from Depo-Provera treated mice were stained for the DC markers

CD11c and MHCII. Depo-Provera treated mice were used because they have

identical hormone conditions which mean that any variation between mice is not due

to them being in different estrous cycle stages. In order to effectively quantify cell

localisation pixel counting was compared with cell counting (Fig 4.2). There were

no significant differences between the two quantification methods.

There was more variation in cellularity between different mice than within one

mouse; some showed low cellularity (Fig 4.3A and B), while others showed high

cellularity (Fig 4.3C and D). There were no significant differences in the numbers of

CD11c+, MHCII+ or CD11c+MHCII+ cells in the epithelia of lower vagina compared

to the upper vagina/cervix (Fig 4.3E-G). There were no significant difference in the

numbers of CD11c+MHCII- and CD11c+MHCII+ cells in the epithelia (Fig 4.3E&G),

but very few MHCII+CD11c- cells in the epithelia (Fig 4.3F).

4.2.3. Localisation of CD68 and MHCII expressing cells throughout

the FRT

To compare macrophage cell numbers in the epithelia at different locations in the

lower FRT sections of FRT tissue from Depo-Provera treated mice were stained for

the tissue macrophage marker CD68 and a marker of antigen presenting capability

MHCII.

The majority of stained cells were in the submucosa with smaller numbers in the

epithelium (Fig 4.4A-D). There was no significant difference between the numbers

of CD68+, MHCII+ or CD68+MHCII+ in the epithelia of the lower vagina or upper

vagina and cervix (Fig 4.4E-G).



31

4.2.4. Localisation of GR-1 and CD11c expressing cells in the FRT

In progesterone treated mice there is infiltration of neutrophils (Fig 3.8). To

investigate if this infiltration has any affect on DCs FRTs from progesterone treated

mice were stained for CD11c and a neutrophil/monocyte marker Gr-1.

GR-1+ staining was absent in tissue sections stained with the isotype control (data

not shown) and was also absent from tissue sections from the estrus stage (data not

shown). There was a great deal of variation within the tissue. There were areas of

very high GR-1 staining (Fig 4.5A and B) and areas of very low Gr-1 staining (Fig

4.5C and D) present within 1 tissue section. There was no significant difference in

the number of CD11c+ cells in the upper vs. lower tract or in the level of neutrophil

infiltration in the outer epithelia. There was no significant difference in the number

of CD11c+ cells in the areas where there were a lot of GR-1+ cells compared to areas

where there were fewer GR-1+ cells.

4.2.5. Immunohistochemistry versus flow cytometry for determining

cell localisation

In Figure 4.6A cells from the pooled, digested vaginal epithelium of mice in diestrus

was stained for CD11c vs. MHCII. 2 populations of CD11c+MHCII+ cells were

seen. Split populations like this cannot be detected using immunohistochemistry;

either because of the higher background fluorescence present in tissue sections (due

to mucus), which would mean low expression is not above the isotype control, or

because immunohistochemistry lacks the sensitivity to differentiate between low,

intermediate and high expressing cells.

Flow cytometry allows staining for more markers (Fig 4.6B and C), which gives a

more comprehensive charactersation of the cell populations present.

Immunohistochemistry, however, is superior for showing localisation of cells within

tissue. Cryopreservation and fixation of tissue is relatively unintrusive and the tissue

and the cells within it remain intact and in place. With flow cytometry the tissue is

processed into a single cell solution, so the validity of assumptions about localisation

depends on the ability to separate the epithelia from the underlying lamina propria.

This is done by manually scraping the epithelia off and using an enzyme mix to



32

break the intercellular bonds. While this can be done to a certain level of accuracy

there is still a much higher chance of cells from the lamina propria becoming washed

onto the epithelium or vice versa. It is also lacks the subtlety to measure cells that

line the border between the basal layer of the epithelia and the submucosa, or cells

that reside in one but extend processes into the other as many of the CD11c+ cells in

the FRT do.

4.3. Discussion

The localisation populations of CD11c+ DCs, CD68+ macrophages and Gr-1+

neutrophils within the FRT of progesterone treated mice were examined. Mice from

the same estrous cycle stage could be used, however (as shown in chapter 3), cyclic

changes are not necessarily consistent throughout the tissue and vaginal smears are

not always clear cut. It is therefore better to use Depo-Provera treatment as a control

for cycle stage variation.

4.3.1. Hormonal effects on cell localisation

CD11c+ cells (predominantly DCs) are the main antigen presenting cell that are

responsible for initiating protective immune responses. There were increased

numbers of CD11c+ cells in the epithelia in metestrus and in progesterone treated

mice. This is consistent with published literature [81-83], which shows hormone

mediated changes in cell recruitment in the vagina. There were differences in

CD11c+ cell localisation in different hormonal conditions in the reproductive tract

(Fig 4.1), with estrus stage showing extremely low levels of CD11c+ cells in the

vaginal epithelia. There is evidence that the number of immune cells fluctuate in the

FRT over the estrous cycle [28, 82-84] and that this may affect immunity and

susceptibility to disease. The localisation and function of DCs and macrophages in

the FRT in rats and humans are affected by sex hormones [21, 56, 59, 81-83].

Langerhans’ cell (LC) numbers vary over the reproductive cycle [21, 83]. In mice,

the repertoire of DC subtypes also changes at different stages of the estrous cycle

[28].

In the gut DCs can reach out into the lumen from beneath the epithelial layer and

sample antigen [85-87] and during infection it has been observed that DCs in the

outer epithelia of the vagina also extend processes into the lumen [88]. At estrus no



33

CD11c+ cells were observed reaching to the luminal edge of the epithelia, whereas at

metestrus CD11c+ cells were seen throughout the epithelial layers, similar to in the

skin, where LCs sit within the stratified cell layers of the epidermis [48, 49, 89]. It

appears that luminal antigen-sampling by CD11c+ cells does not occur at estrus,

which may have implications for antigen presentation and initiation of the immune

response.

CD68+ tissue macrophages are professional phagocytes that perform homeostatic

roles within the tissue , including clearance of apoptotic cells and would repair [90,

91]. Macrophages have a homeostatic role in uterus and ovary in humans [40].

Vaginal macrophages are phenotypically different to macrophages in other mucosal

sites [92] and peripheral blood [93]. CD68+ tissue macrophages can express MHCII

and present antigen to T cells [94] and can also influence the initiation of immune

responses by cross-talk with other APCs and T cells [90]. Macrophages express the

estrogen receptor and are responsive to progesterone, with macrophage numbers in

the endometrium fluctuating over the menstrual cycle [40]. No difference in CD68+

cell localisation was observed in the murine vagina under different hormonal

conditions, implying that there would be no difference in vaginal macrophage

antigen presentation during the estrous cycle.

4.3.2. Distribution of cells along the lower FRT

In humans immune cells are not distributed throughout the lower FRT uniformly.

DCs are most abundant in the cervix and other immune cells are clustered around the

cervical transformation zone, where the ecto- and endocervix meet, with very few

cells in the vagina [95]. Comparisons of the proximal and distal vagina have shown

some differences in T cell distribution [95]. Differences in cell localization and

numbers along the tract may affect immunity. Different pathogens invade the FRT

at different locations. For example, the vagina is prone to Candida albicans and

Trichomonas vaginalis infection, the cervix is prone to Chlamydia trachomatis and

Neisseria gonorrhoea infection and HPV preferentially infects the transformation

zone [28, 47, 95].

There was variation in the localisation of macrophages, DCs and, neutrophils in the

FRTs of progesterone treated mice (Fig 4.3 and 4.4), leading to lots of distinct



34

microenvironments along the length of the lower reproductive tract. However, no

statistically significant pattern was observed, that would account for the consistent

differences in susceptibility to infection described above.

4.3.3. Affects of neutrophils on APCs

Neutrophils enter the FRT from the blood in response to IL-8 and other chemokines

produced by epithelial cells [40]. Infection and insemination results in increased

neutrophil recruitment to the FRT in humans [40]. In humans rapid neutrophil

recruitment in response to stimuli is particularly associated with the cervix [40]. In

the human uterus falling progesterone levels triggers an increase in IL-8 production

by uterine epithelial cells [40].

The high progesterone stages of estrous cycle are characterized by physiological

infiltration of leukocytes, the majority of which are neutrophils (Fig 3.3, 6 and 7)

[40, 82]. GR-1 is expressed in high levels on neutrophils, although, it is also

expressed by plasmacytoid DCs and some monocytes [96]. There was variation in

the density of GR-1+ staining in the FRTs of progesterone treated mice (Fig 4.5),

however no pattern in distribution of GR-1+ cells in different areas of the lower FRT

were observed. The differences are most likely due to variations in the

microenvironments over the epithelium caused by the normal undulations in the

tissue.

Neutrophils have been shown to effect APC recruitment and function and may

interact directly with APCs [25, 97-101]. To investigate the affect of neutrophil

infiltration to the epithelia on CD11c+ cell localization in the epithelia tissue sections

from the vaginas and cervixes of Depo-Provera treated mice were stained for Gr-1

and CD11c. There was no significant difference in the number of CD11c+ cells in

the areas where there were a lot of GR-1+ cells compared to areas where there were

fewer GR-1+ cells. This does not exclude the possibility that GR-1+ cells were

responsible for recruitment of CD11c+ cells (Gr-1+ cells were present in all sections

with CD11c+ cells), but shows that increasing numbers of GR-1+ cells did not cause

a proportional increase in CD11c+ cells. There was no correlation between different

density of GR-1+ staining and CD11c+ staining. This shows that while hormonal

changes mediate changes in neutrophil infiltration (Fig 3.6-8), the frequency of



35

neutrophils does not affect DC localisation in the epithelia. This does not rule out

the possibility that the presence of neutrophils may affect DC localisation; just that it

is not ‘dose dependant.’

4.3.4. Quantification of confocal images

While no significant differences in the 2 quantification methods were seen, there are

still issues with cell counts and pixel counts. Due to the changes in the epithelia

thickness measurement of cells per unit area and pixel counts expressed as a

percentage of total pixels (essentially an expression of area) are flawed. For example

if the cell numbers in the epithelia remain the same but the epithelia thickens this

analysis would show a reduction in cells/mm2 despite unchanging cell numbers.

Measuring the length of the epithelium along the basal edge, which remains

relatively unchanged over the estrous cycle (as opposed to measuring the length of

the luminal edge of the epithelium, which becomes disrupted), allows for a more

accurate prediction of the relative number of cells at different locations and at

different cycle stages. Pixel counts cannot discriminate between the same number of

cells with greater surface expression and therefore more staining and increased

numbers of cells expressing similar levels of surface protein. Both of these scenarios

would result in an increased pixel count.



36

Figure 4.1: Comparison of DC and macrophage localisation within the epithelia

of the FRT at different stages of the estrous cycle.

7 µm sections from CD45.1 or C57BL6 mice were stained for CD11c or CD68

(green) and DAPI (blue). Images are representative of 3-6 images per mouse.

Estrous cycle stage was determined by H&E staining of tissue sections. A) CD11c

stained vagina at estrous. B) CD68 stained vagina at estrus. C) CD11c stained vagina

at metestrus D) CD68 stained vagina at metestrus. White arrows show positively

stained cells. Ep. Epithelia L. Lumen SM. Submucosa.



37

Figure 4.2: Pixel count versus cell count analysis.

A) For pixel counts the analysis was done in Photoshop. A line was drawn around

the epithelium and then the epithelium was selected. A total pixel count for the

epithelium was performed followed by selected counts on the green, red and yellow

channels in turn. B) The epithelium was then excluded and the submucosa selected.

Total pixel counts for the submucosa and for each individual channels was

performed. C) For cell counts the length of the basal layer of the epithelia was

drawn using the LSM image browser software. D) Cell counts were performed using

ImageJ software. E) For the total CD11c counts the red and yellow channels were

added together and expressed as a percentage of total pixels for each field of view.

Percent pixel values from 3-6 fields of view were averaged to give a single point for

the lower and upper vagina for each mouse. F) Cell counts were expressed as

cells/mm. Each point represents the average cells/mm from 3-5 fields of view. G)

To compare the 2 methods of quantification cells/mm was plotted by % total pixels.



38

Figure 4.3: Comparison of CD11c+ cells in the epithelia and sub mucosa of the

vagina and upper vagina/cervix of Depo-Provera treated mice.

4 female CD45.1 mice were injected subcutaneously with 100 µl of 30 mg/ml Depo-

Provera. 5 days later reproductive tracts were removed and 7 µm sections were cut,

then stained for CD11c (red), MHCII (green) and DAPI (blue). Representative

images of the lower vagina (A and C) and upper vagina/cervix (B and D) of 2

individual mice. A) and B) Mouse with comparatively low numbers of cells in the

FRT. C) and D) Mouse with comparatively high numbers of cells in the FRT. (E-G)

Average cells/mm CD11c single positive (E), MHCII single positive (F) and CD11c

and MHCII double positive (G) cells per mouse. Images are representative of 3-6



39

images lower vagina and 3-6 images per upper vagina/cervix from 4 different mice.

Ep epithelia, L lumen, SM submucosa

Figure 4.4: CD68 and MHCII positive cells in the epithelia of Depo-Provera

treated mice.

Vaginas from 4 B6.CD45.1 mice treated with 100 µl 30 mg/ml Depo-Provera were

cut into 7 µm tissue sections. A-D) Representative images of CD68 (red), MHCII

(green) and DAPI (blue) stained sections of upper and lower vagina. E-G)

Comparison of the average number of CD68+, MHCII+ and CD68+MHCII+ double

positive cells per mm in the epithelia of the upper and lower vagina. Images are

representative of ≥5 images per group from 4 different mice. Ep epithelia, L lumen,

SM submucosa



40

Figure 4.5: DCs and neutrophils in the vagina and cervix of Depo-Provera

treated mice.

The vagina and cervix of 4 Depo-Provera treated B6.CD45.1 mice were isolated and

7 μm tissue sections were counterstained with DAPI (blue) and stained with Gr-1

(red) and CD11c (green). A-D) Images are split into from top left to bottom right -

blue only, red only, green only and merged image. A) and B) Vagina and cervix of 1

mouse. C) and D) Vagina and cervix of a 2nd mouse. E) CD11c+ cell and Gr-1+ cell

interacting within the epithelia. F) Higher magnification z stack image of E). G)

CD11c+ cells numbers in the epithelia of upper and lower vagina. The graph

compares DC numbers in areas of high neutrophil density (A and B) and low



41

neutrophil density (C and D). Each point represents one field of view. Images are

representative of ≥3 images per group from 4 different mice. Ep epithelia, L lumen,

SM submucosa

Figure 4.6 Diverse phenotypes of CD11c+MHCII+ cells from the vaginal

epithelia.

C57BL/6 female mice. Dead cells were excluded using a live/dead discriminator.

Live cells were selected based on forward scatter vs. side scatter. Haematopoietic

cells were selected based on CD45.2 expression. A) CD11c vs. MHCII expression

in CD45.2+ cells. B) F4/80 vs. CD11b expression in CD45.2+CD11chiMHCII+ cells.

C) F4/80 vs. CD11b expression of CD45.2+CD11cloMHCII+ cells.



42

5. Characterisation of the lymph nodes draining the

reproductive tract

5.1. Introduction

The immune system at mucosal sites is considered to be linked, because of the

preferential trafficking of T cells primed in mucosal dLNs to other mucosal

lymphoid tissues rather than to the spleen or peripheral LNs [102, 103]. This means

that immune responses primed at one mucosal site can elicit protective immunity at

distal mucosal sites.

Mucosal lymph nodes have a different ontogeny to other peripheral LNs and are the

first secondary LNs to develop in the embryo [104]. In most cases, mesenchymal

cells initiate LN development by production of CXCL13, which binds CXCR5 on

LTi precursor cells. LTβ signalling between mesenchymal cells and LTi cells causes

differentiation of the mesenchymal cells into stromal organiser cells, which give rise

to the LN stromal cell populations (MRCs, FRCs etc) [104, 105]. Hence, mice

deficient in CXCR5 and CXCL13 lack iliac LNs and other peripheral LNs [106] and

LTβ deficient mice lack most peripheral LNs and Peyer’s Patches [69, 104, 107,

108]. Surprisingly, MLNs and CLNs are found in mice lacking CXCR5 and

CXCL13 [106]. Similarly, MLNs, CLNs and ILNs are present in LTβ deficient mice

[69, 104, 107, 108]. These findings imply that mucosal LNs associated with the

airways and gut are fundamentally different to other secondary lymphoid tissues.

Lymph nodes function as local hubs for interactions between immune cells. Antigen

presentation to naïve T cells as well as T-B cell interactions, which lead to initiation

of the cellular and humoural adaptive response, occur here [28, 36, 44, 109-116].

The environment in draining lymph nodes can affect T cell responses [44, 112].

Upon activation naïve T cells change their expression of surface adhesion molecules,

which allows them to exit the LN via the lymphatic vessels. The environment

produced by APCs and stromal cells in LNs induces expression of different adhesion

molecules on T cells [36, 44, 117]. The endothelial cells in different mucosal tissues

and lymph nodes express different adhesion molecules meaning that only certain

subsets of T cells can enter [69, 71, 84, 118]. For example, MLN stromal cells



43

produce retinoic acid (RA), which promotes gut-homing by inducing expression of

the mucosal addressin MAdCAM-1 [36-38, 44, 114]. Lymph node stromal cells

have also been shown to present antigen [119]. There are variations in LN stromal

cells in different LNs [36-38], which means that there may be LN specific

differences in antigen presentation intrinsic to the stromal cells. In this chapter, the

architecture and cellular composition of LNs draining the FRT were compared with

other mucosal (mesenteric and cervical) and skin draining LNs.

5.2. Results

5.2.1. Iliac lymph nodes are architecturally similar to other mucosal

lymph nodes

While lymph node architecture has been well characterised [34, 114, 115, 120] there

has been little comparison of different lymph nodes. Differences in architecture of

the LNs especially the locations of APCs and T cells may point to differences in

antigen-presentation and subsequent immune responses. Therefore, the

macrostructure of the cervical, mesenteric and iliac lymph nodes were compared.

Mice expressing a variety of fluorochromes under the control of cell-specific

promoters/locus control regions allow visualisation of lymph nodes in steady state

without the need to treat with injectable markers.

CD19CreR26REYFPVaDsRedxC57xC57 mice, which express EYFP in CD19

positive cells and DsRed in CD2 positive cells, were dissected on a fluorescent

stereo microscope to show the locations of cervical, mesenteric and iliac lymph

nodes (Fig 5.1B). Lymph nodes have discreet T cell (depicted blue after image

analysis) and B cell (green) zones. There were differences in size of lymph nodes

with MLNs being largest. The superficial cervical lymph nodes are also relatively

large with smaller deep cervical LNs. Iliac lymph nodes are small and are slightly

inconsistent in location (data not shown). They are sometimes both on one side of

the aorta and sometimes on both sides, but are always just above the base of the

spine. It is unknown if this variability could affect immune responses in different

animals.

Sections of frozen iliac, mesenteric and cervical lymph nodes from naïve C57BL/6

mice were stained with markers for DCs (CD11c+, MHCII+), B cells (MHCII+),



44

marginal reticular cells (gp38+) and lymphatic (CD31+, gp38+) and vascular

endothelial cells (CD31+). Fig 5.2 shows representative images of ILNs, CLNs and

MLNs. MHCII +CD11c- B cells are situated in the cortex in defined follicles.

CD11c+ DCs are located in the medulla and surround the B cell follicles but are

rarely seen within the B cell follicles. CD31+gp38- HEVs are located in the medulla.

The gp38+ MRCs form structural support for the T cell zones and subcapsular sinus

(SCS) [34, 117]. The localisation of CD11c+ cells around CD31+ vessels, in the T

zone and around B cell follicles were comparable, as was the number of B cell

follicles (Fig 5.2).

5.2.2. Iliac lymph nodes have comparable antigen-presenting cell

subpopulations to cervical lymph nodes, but not mesenteric lymph

nodes

Differences in DC numbers and phenotypes in the LN can affect downstream

immune responses. The DC populations in iliac LNs were characterised by their

expression of surface markers and compared with other mucosal and peripheral LNs.

LNs from Depo-Provera treated mice were pooled into groups of cervical lymph

nodes, iliac lymph nodes, mesenteric lymph nodes and a group of ‘other’ lymph

nodes, which included axillary and inguinal lymph nodes and then stained for

CD11c, MHCII, CD103 and CD11b.

CD11c and MHCII double positive cells were split into 2 subpopulations:

CD11chiMHCII+ and CD11c+MHCIIhi cells (Fig 5.3C) [121, 122]. The percentage

of DCs was consistently low, with total CD11chiMHCII+ and CD11c+MHCIIhi cells

accounting for less than 1.5% of total live LN cells in all lymph nodes. Comparison

of different lymph nodes showed different percentages of CD11chiMHCII+ and

CD11c+MHCIIhi cells (Fig 3C and D). Iliac lymph nodes had a much lower

percentage of both APC populations compared to other lymph nodes.

Both Iliac LNs and mesenteric LNs had lower percentages of CD11chiMHCII+

antigen presenting cells compared to cervical LNs and the pool of other lymph

nodes. Based on their CD103 and CD11b expression DCs could be split into 3

subpopulations (Fig 5.3C): CD103loCD11b+ cells, CD103+CD11b+ cells and

CD103+CD11b- cells. In all lymph nodes CD103loCD11b+ cells were the largest



45

population. Percentages of CD103loCD11b+, CD103hiCD11bhi and CD103hiCD11blo

subtypes of CD11chiMHCII+ and CD11c+MHCIIhi cells were similar in CLNs, ILNs

and other LNs, but MLNs show higher CD103hiCD11bhi and CD103hiCD11blo cells.

In ILNs, CLNs and OLNs CD103loCD11b+ cells dominated. In the MLNs there

were similar percentages of all 3 subtypes with a roughly 3-fold increase in the

CD103+ subsets.

5.2.3. Comparatively low RALDH expression in iliac lymph nodes

To compare RALDH expression of both haematopoietic and non-haematopoietic

cells in cervical and iliac lymph nodes pooled LN cells from 5 Depo-Provera treated

C57BL6 mice each were stained for RALDH expression and surface molecules.

There were almost no detectable RALDH+CD45.1- non-haematopoietic cells in any

LNs (Fig 5.4B). The highest found was 0.7% of the CD45.1- population (or 21

cells), which is below the reliable detection of this assay. Iliac lymph nodes had the

lowest percentage of RALDH+CD45.1+ cells with roughly equal numbers in the

CLNs and ‘other’ LNs. A lower percentage of iliac lymph node CD45.2+ cells

showed RALDH activity compared to cervical lymph nodes. Fewer iliac lymph

node CD11c+MHCII+ cells show RALDH activity compared to cervical lymph node

cells.

5.3. Discussion

The FRT is a mucosal tissue, however in the lower FRT the epithelia and the

resident immune cells bear similarities to the skin. To determine if the FRT dLNs

were more similar to mucosal LNs or peripheral LNs that drain the skin, the iliac

LNs were compared to other LNs.

Iliac LNs are architecturally similar to MLNs and CLNs. ILNs have a lower

percentage of CD11c+MHCII+ APCs than MLNs or CLNs, even after accounting for

the lower total cell numbers (Fig 5.3). Different ratios or numbers of DC subtypes

may imply that there are functional differences between lymph nodes.CD11c+ DCs

in both the tissues and lymph nodes can be divided into functionally different

subtypes based on the expression of surface molecules. DC subtypes have been well

characterised in many other tissues [26, 50, 110, 112, 113, 123-128]. In the lungs



46

CD11c+MHCIIhi cells, but not CD11chiMHCII+ have been shown to be responsible

for antigen transport to the dLN [122]. CD11c+MHCIIhi cells are thought to be

migratory DCs while CD11chiMHCII+ cells are LN resident DCs [128]. CD103+

DCs have been best characterised in the gut [85, 110, 129], but are also found in

other lymph nodes [113, 130]. CD103 expression is involved in attachment to

mucosal epithelial cells and is found in migratory DCs from mucosal lymph nodes

[85, 110, 131]. They have been shown to promote tolerogenic responses and effector

T cell homing to mucosal tissues [110, 129, 131]. ILNs have similar APC

populations to CLNs, but not MLNs.

The vitamin A metabolite retinoic acid (RA) is linked to oral tolerance because it can

affect T cell maturation by promoting the differentiation of Tregs and gut homing

[37, 113]. Retinal is converted to RA by RALDH enzyme activity [132]. In the

MLNs tissue-derived DCs and macrophages [133], as well as lymph-node resident

non-haematopoietic stromal cells [36-38], express RALDH and are thought to be

important for oral tolerance induction [37, 38, 114]. RALDH expression is

associated with the induction of gut-homing T and B cells and tolerance [37, 38, 113,

133, 134]. RALDH is also expressed by CD11c+MHCII+ cells in other mucosal

lymph nodes such as the mediastinal LNs that drain the lung, but is expressed at

much lower levels in skin-draining LNs and in the spleen [113]. In this study

RALDH expression in iliac LNs was measured and compared to other LNs, which

has not been investigated before. While RALDH expression was lower in iliac LNs

compared to other peripheral LNs, there is some evidence that RA may be important

in immunity in the FRT because there is increased HIV shedding in women with

vitamin A deficiencies [60].

The lower RALDH expression and lower DC ratios implies that ILNs are dissimilar

to other mucosal LNs. These differences in APC populations and the environment

within the ILNs potentially means that antigen presentation and the subsequent

immune response and long term immunity initiated will be different to the immune

responses initiated with the same antigen at a different mucosal site. This has

implications for treatment and prevention of STIs and for vaccination strategies

administered intravaginally or for vaccines that need to induce immunity in the FRT,

but are administered at different sites.



47

Figure 5.1: Location of mucosal lymph nodes in mice.

A) Nomenclature of lymph nodes in mice, from [120]. B)

CD19CreR26REYFPVaDsRedxC57xC57 mouse was imaged. Blue colouration

shows T cells and green shows B cells. Arrows show trachea (red), small intestine

(yellow), colon (blue) and uterine horns (white). The zoomed images are separate

higher magnification images. Data is representative of 2 mice.

1. Mandibular LN2. Accessory mandicular LN3. Superficial parotid LN4. Cranial deep cervical LN5. Proper axillary LN6. Accessory axillary LN7. Subiliac LN8. Sciatic LN9. Popliteal LN10. Cranial mediastinal LN11. Tracheobronchal LN12. Caudal mediastinal LN13. Gastric LN14. Pancreaticoduodenal LN15. Jejunal LN16. Colic LN17. Caudal mesenteric LN18. Renal LN19. Lumbar aortic LN20. Lateral iliac LN21. Medial iliac LN22. External iliac



48

Figure 5.2: Architecture of mucosal lymph nodes.

3 C57BL/6 mice were injected subcutaneously with 100 µl 30 mg/ml Depo-Provera.

5 days later estrous cycle arrest was confirmed by vaginal washes. Lymph nodes

were removed and frozen in OCT. 8 μm sections were cut on a cryostat and then

stained for MHCII (blue), gp38 (green), CD31 (yellow) and CD11c (red). Dashed

lines show B cell follicles. Yellow arrows show HEVs. Overlapping images were

taken on a confocal microscope and stitched together using Fiji software.



49

Figure 5.3: Comparison of DC subtypes in iliac, mucosal and peripheral lymph

nodes.

Lymph nodes from 5 female Depo-Provera treated C57BL/6 mice were pooled and

stained for surface markers. A) Gating strategy. Live cells are selected based on

forward scatter vs. side scatter. Doublet discrimination based on pulse width vs.

forward scatter. Both CD11chiMHCII+ and CD11c+MHCIIhi antigen presenting cells

are analysed. B) Isotype control. C) CD103 and CD11b expression of

CD11chiMHCII+ and CD11c+MHCIIhi antigen presenting cells. D) % of total cells



50

that are CD11chiMHCII+ and CD11c+MHCIIhi in different lymph nodes. E) % of

total gate of 3 subpopulations of CD11bCD103 expressing cells. 1) CD11chiMHCII+

2) CD11c+MHCIIhi i) CD103loCD11b+ ii) CD103hiCD11bhi iii) CD103hiCD11blo

Figure 5.4: RALDH expression in iliac and peripheral lymph nodes.

LNs from 5 female Depo-Provera treated mice were pooled into groups of CLNs,

ILNs and OLNs and stained for CD45.2 and with ALDEFLUOR for RALDH

activity. A) Gating strategy. Cells were gated on based on forward scatter vs. side

scatter. Doublets were excluded based on pulse width vs. forward scatter and dead

cells were excluded based on positive staining for the live/dead cell marker. B)

Cells were assessed for RALDH activity with or without the DEAB inhibitors.

DEAB inhibitors prevent RALDH activity. C) RALDH+ cells as a percentage of

total cells.



51

6. Discussion

6.1. General conclusions

The FRT undergoes drastic changes over the reproductive cycle. In progesterone

high stages there is a large influx of leukocytes. The distribution of APCs is

constant along the length of the vagina and cervix. The numbers of Gr-1+ cells

varies greatly over the course of the reproductive cycle. The distribution of Gr-1+

cells within the tissue varies, with some areas showing relatively high numbers of

neutrophils in the outer epithelia and some areas showing much lower infiltration.

This study found no significant pattern to the areas of low or high neutrophil influx.

It is possible that this represents individual differences in the architecture of the

epithelia caused by differences in epithelial folding, thickness and mucus flow. This

shows that the frequency of APCs is constant throughout the tract, but that the local

environments within the tract may have differences in antigen capture and

presentation. There is a lot of variation along the length of the lower reproductive

tract and between mice, which may affect the design of future experiments. For

example more mice may be needed to get statistically significant results.

There is some evidence that neutrophils in the human fallopian tube [40, 135] are

phenotypically different to normal blood neutrophils. In mice anti-Gr-1+ cell

depletion results in estrous cycle arrest [136], implying a physiological role for

neutrophils. However, monocytes also express Gr-1 and depletion with the

neutrophil specific anti-Ly6G antibodies does not result in estrous cycle arrest

(Schäfer unpublished observations). It is not known if neutrophils recruited to the

vagina in the murine estrous cycle are different to blood neutrophils. There is a

rapidly growing literature on newly characterised neutrophil phenotypes and

functions [25, 97-101, 135]. This has led to a growing appreciation that neutrophils

are not merely rapidly recruited killer cells, but may have varied roles in the immune

response. [99, 137]

Iliac lymph nodes are architecturally similar to mesenteric and cervical nodes in

normal mice, as judged by distribution of B cell follicles, vascular and lymph

endothelia and DC localisation. Hence, there is unlikely to be any major differences



52

in the cellular organisation or trafficking between iliac and other mucosal lymph

nodes. This does not exclude the possibility of differences in architecture occurring

during an immune response or as a result of cyclic changes.

The lymph nodes draining the FRT have a lower percentage of APCs compared to

other mucosal lymph nodes. APC’s can be divided into subpopulations based on

their expression of the surface molecules CD103 and CD11b. The ratios of different

subpopulations in FRT dLNs are comparable to cervical lymph nodes, but are

different to MLN ratios. In MLNs there are higher percentages of the CD103hi

subtypes. CD103 is involved in binding to epithelial cells so CD103 expression is

associated with cells that have migrated from the epithelia of mucosal tissues [138].

This implies that there are either fewer migratory DCs coming from the FRT and

nasal tissue or that CD103+ represent only a minor subpopulation of the migratory

DCs from these sites.

RALDH expression is lower in iliac LNs compared to cervical and other peripheral

LNs. Low RALDH expression implies a bias away from tolerogenic responses and

trafficking to mucosal tissues. However, tolerogenic responses in the skin are

associated with increased vitamin D metabolites and not RA [134], so it is possible

that vitamin D could induce tolerogenic responses in the FRT. Vitamin A deficiency

is associated with susceptibility to HIV infection [60], but the implication of this is

unclear. One possibility is that T cells primed in mucosal lymph nodes do not traffic

to mucosal tissues because they have not upregulated MAdCAM-1 and would traffic

to other LNs.

DCs represent a lower percentage of iliac LNs than in other LNs, this could mean

that the FRT has a lower potential for stimulating immune responses. This may be

due to either or both lower LN resident DCs and lower migratory DCs. Since the

ratios of DC subpopulations are equivalent to other lymph nodes, the data in this

study would imply an overall drop in DC numbers rather than changes in specific

subpopulations. If antigen presentation is equivalent in ILNs this could mean that

another cell type is providing antigen presentation, for example macrophages.



53

6.2. Disadvantages of Mouse Models

The observation of a large infiltration of neutrophils during metestrus/diestrus and in

progesterone treated animals raises questions about the appropriateness of this model

to study human FRT biology. Neutrophils are not seen in large numbers in humans

either during the normal menstrual cycle or after hormone contraceptive treatment

[40]. There is no increase in neutrophils during the progesterone high stage of the

menstrual cycle despite increases in neutrophils chemoattractants [40, 139].

Mouse models have advantages over doing work on humans or macaques. Firstly,

there are lower ethical restraints. Secondly, there is a greater potential for genetic

manipulation due to the diverse genetic models in mice. Whilst not exactly the same

as humans, mice provide a good model for proof of concept studies.

An alternative would be to use in vitro experiments based on human FRT cell lines

or explants. Due to the high turnover of immune cells coming into the tissue from

the blood and that many of the important immunological events take place in the

lymph node, these methods come with a lot of caveats and cannot effectively mimic

complex inter-system interactions involved in immunity.

6.3. Future Work

Phenotyping of the neutrophils of the lower FRT and comparison with both blood

and uterine neutrophils would shed some light onto the homeostatic function of

neutrophils. It would also be interesting to compare the early events in primary

infection with or without neutrophils present to see if they have any effect on APCs

or the initiation of the adaptive immune response.

There is some evidence that macrophages in the human FRT are phenotypically

different to other macrophages [92] and it would be interesting to see what

differences this would have on immunity and if similar differences were also

observed in mouse models. FRT DC phenotypes have been more thoroughly studied

in mice [81].

‘Mucosalness’ of LNs is difficult to define. Immune responses to administered

antigen may prove a better way of comparing mucosal lymph nodes. It would be

important to study the trafficking of T cells primed in the FRT and their expression



54

of MAdCAM-1 and other tissue addressins. This would have important implications

for systemic immunity and vaccine design. The data produced here provides a

foundation for such studies.



55

Abbreviations

AIDS acquired immunodeficiency syndrome

APC antigen-presenting cell

APC allophycocyanin (Materials and Methods only)

BSA bovine serum albumin

CLN cervical lymph node

Cy cyanin

DAPI 4',6'-diamidino-2-phenylindole

DC dendritic cell

DEAB diethylaminobenzaldehyde

dLN draining lymph node

FACS fluorescence-activated cell sorting

FCS Foetal calf serum

Flt-3L fms-like tyrosine kinase-3 ligand

FRC fibroblastic reticular cell

FRT female reproductive tract

FSH follicle-stimulating hormone

gp38 podoplanin

GM-CSF granulocyte-macrophage colony stimulating factor

Gr-1 granulocyte differentiation antigen 1

H&E haematoxylin and eosin

HEV high endothelial venule



56

HIV human immunodeficiency virus

HPV human papilloma virus

HSV herpes simplex virus

IL interleukin

ILN iliac lymph node

LC Langerhans’ cells

LH luteinizing hormone

LN lymph node

LPS lipopolysaccharide

LT lymphotoxin

LTi lymphoid tissue inducer

MAdCAM-1 mucosal addressin cell adhesion molecule-1

MCP-1 monocyte chemoattractant protein -1

MHCII major histocompatibility complex type two

MIP-2 macrophage inflammatory protein 2

MLN mesenteric lymph node

MRC marginal reticular cell

OCT optimal cutting temperature

OLN ‘other’ (mixture of axillary and inguinal) lymph nodes

PAMP pathogen-associated molecular pattern

PBS phosphate buffered saline

PE phycoerythrin



57

PerCP peridinin chlorophyll A protein

PFA paraformaldehyde

PP Peyer’s patch

PRR pathogen-recognition receptor

RA retinoic acid

RALDH retinaldehyde dehydrogenase

SC stratum corneum

SCS subcapsular sinus

STI sexually transmitted infection

TGF transforming growth factor

Th1 type 1 helper T cell

TLR toll-like receptor

Treg regulatory helper T cell



58

References

1. Starnbach MN, Roan NR. Conquering sexually transmitted diseases. Nat RevImmunol 2008,8:313-317.

2. Iwasaki A. Antiviral immune responses in the genital tract: clues for vaccines. NatRev Immunol 2010,10:699-711.

3. WHO. GLOBAL PREVALENCE AND INCIDENCE OF SELECTED CURABLE SEXUALLYTRANSMITTED INFECTIONS OVERVIEW AND ESTIMATES. 2001.

4. UNAIDS Report on the Global AIDS Epidemic. 2010.5. Bosch FX, Lorincz A, Muñoz N, Meijer CJLM, Shah KV. The causal relation between

human papillomavirus and cervical cancer. Journal of Clinical Pathology2002,55:244-265.

6. Health Protection Agency, UK. Health Protection Report 2010,4.7. Holmgren J, Czerkinsky C. Mucosal immunity and vaccines. Nat Med 2005.8. Mestecky J, Raska M, Novak J, Alexander RC, Moldoveanu Z. Antibody-mediated

protection and the mucosal immune system of the genital tract: relevance tovaccine design. Journal of Reproductive Immunology 2010,85:81-85.

9. Nagler-Anderson C. Man the barrier! strategic defences in the intestinal mucosa.Nat Rev Immunol 2001,1:59-67.

10. Ogra PL, Faden H, Welliver RC. Vaccination Strategies for Mucosal ImmuneResponses. Clin. Microbiol. Rev. 2001,14:430-445.

11. Artis D. Epithelial-cell recognition of commensal bacteria and maintenance ofimmune homeostasis in the gut. Nat Rev Immunol 2008,8:411-420.

12. Turner JR. Intestinal mucosal barrier function in health and disease. Nat RevImmunol 2009,9:799-809.

13. Mowat AM. Anatomical basis of tolerance and immunity to intestinal antigens.Nature Reviews Immunology 2003,3:331-341.

14. Doss M, White MR, Tecle T, Hartshorn KL. Human defensins and LL-37 in mucosalimmunity. Journal of Leukocyte Biology 2010,87:79-92.

15. Malm J, Sorensen O, Persson T, Frohm-Nilsson M, Johansson B, Bjartell A, et al. TheHuman Cationic Antimicrobial Protein (hCAP-18) Is Expressed in the Epithelium ofHuman Epididymis, Is Present in Seminal Plasma at High Concentrations, and IsAttached to Spermatozoa. Infect. Immun. 2000,68:4297-4302.

16. Jones RE, Lopez KH. Human reproductive biology. GB: Academic Press; 2006.17. Vareille M, Kieninger E, Edwards MR, Regamey N. The Airway Epithelium: Soldier in

the Fight against Respiratory Viruses. Clin. Microbiol. Rev. 2011,24:210-229.18. Resta SC. Effects of probiotics and commensals on intestinal epithelial physiology:

implications for nutrient handling. The Journal of Physiology 2009,587:4169-4174.19. Wells JM, Rossi O, Meijerink M, van Baarlen P. Epithelial crosstalk at the

microbiota–mucosal interface. Proceedings of the National Academy of Sciences2011,108:4607-4614.

20. Klebanoff SJ, Hillier SL, Eschenbach DA, Waltersdorph AM. Control of the MicrobialFlora of the Vagina by H202-Generating Lactobacilli. Journal of Infectious Diseases1991,164:94-100.

21. Hel Z, Stringer E, Mestecky J. Sex Steroid Hormones, Hormonal Contraception, andthe Immunobiology of Human Immunodeficiency Virus-1 Infection. Endocr Rev2010,31:79-97.

22. Fidel PL, Jr., Cutright J, Steele C. Effects of Reproductive Hormones on ExperimentalVaginal Candidiasis. Infect. Immun. 2000,68:651-657.



59

23. Rodriguez-Pinto D. B cells as antigen presenting cells. Cellular Immunology2005,238:67-75.

24. Weenink SM, Gautam AM. Antigen presentation by MHC class II molecules.Immunol Cell Biol 1997,75:69-81.

25. van Gisbergen KPJM, Geijtenbeek TBH, van Kooyk Y. Close encounters ofneutrophils and DCs. Trends in Immunology 2005,26:626-631.

26. Pulendran B. Division of labor and cooperation between dendritic cells. NatImmunol 2006,7:699-700.

27. Pulendran B, Tang H, Denning TL. Division of labor, plasticity, and crosstalkbetween dendritic cell subsets. Current Opinion in Immunology 2008,20:61-67.

28. Iwasaki A. Mucosal Dendritic Cells. Annual Review of Immunology 2007,25:381-418.29. Shaykhiev R, Bals R. Interactions between epithelial cells and leukocytes in

immunity and tissue homeostasis. Journal of Leukocyte Biology 2007,82:1-15.30. Fritz JH, Le Bourhis L, Magalhaes JG, Philpott DJ. Innate immune recognition at the

epithelial barrier drives adaptive immunity: APCs take the back seat. TrendsImmunol 2008,29:41-49.

31. Swamy M, Jamora C, Havran W, Hayday A. Epithelial decision makers: in search ofthe 'epimmunome'. Nat Immunol 2010,11:656-665.

32. Kato A, Schleimer RP. Beyond inflammation: airway epithelial cells are at theinterface of innate and adaptive immunity. Curr Opin Immunol 2007,19:711-720.

33. Weindl G, Wagener J, Schaller M. Epithelial cells and innate antifungal defense. JDent Res 2010,89:666-675.

34. Mueller SN, Germain RN. Stromal cell contributions to the homeostasis andfunctionality of the immune system. Nat Rev Immunol 2009,9:618-629.

35. Tokoyoda K, Hauser AE, Nakayama T, Radbruch A. Organization of immunologicalmemory by bone marrow stroma. Nat Rev Immunol 2010,10:193-200.

36. Ahrendt M, Hammerschmidt SI, Pabst O, Pabst R, Bode U. Stromal cells conferlymph node-specific properties by shaping a unique microenvironment influencinglocal immune responses. Journal of Immunology 2008,181:1898-1907.

37. Hammerschmidt SI, Ahrendt M, Bode U, Wahl B, Kremmer E, Forster R, et al.Stromal mesenteric lymph node cells are essential for the generation of gut-homing T cells in vivo. Journal of Experimental Medicine 2008,205:2483-2490.

38. Pabst O, Wahl B, Bernhardt G, Hammerschmidt S. Mesenteric lymph node stromacells in the generation of intestinal immune responses. Journal of MolecularMedicine-Jmm 2009,87:945-951.

39. Miller L, Modlin R. Toll-like receptors in the skin. Seminars in Immunopathology2007,29:15-26.

40. Wira CR, Fahey JV, Sentman CL, Pioli PA, Shen L. Innate and adaptive immunity infemale genital tract: cellular responses and interactions. Immunological Reviews2005,206:306-335.

41. Wira CR, Grant-Tschudy KS, Crane-Godreau MA. Epithelial Cells in the FemaleReproductive Tract: a Central Role as Sentinels of Immune Protection. AmericanJournal of Reproductive Immunology 2005,53:65-76.

42. Hammad H, Lambrecht BN. Dendritic cells and epithelial cells: linking innate andadaptive immunity in asthma. Nat Rev Immunol 2008,8:193-204.

43. Haraldsen G, Balogh J, Pollheimer J, Sponheim J, Küchler AM. Interleukin-33 -cytokine of dual function or novel alarmin? Trends in Immunology 2009,30:227-233.

44. Kraal G, Samsom JN, Mebius RE. The importance of regional lymph nodes formucosal tolerance. Immunological Reviews 2006,213:119-130.



60

45. Hladik F, McElrath MJ. Setting the stage: host invasion by HIV. Nat Rev Immunol2008,8:447-457.

46. Iwasaki A. Mucosal dendritic cells. Annual Review of Immunology 2007,25:381-418.47. Quayle AJ. The innate and early immune response to pathogen challenge in the

female genital tract and the pivotal role of epithelial cells. Journal of ReproductiveImmunology,57:61-79.

48. Romani N, Clausen BE, Stoitzner P. Langerhans cells and more: langerin-expressingdendritic cell subsets in the skin. Immunological Reviews 2010,234:120-141.

49. Merad M, Ginhoux F, Collin M. Origin, homeostasis and function of Langerhanscells and other langerin-expressing dendritic cells. Nat Rev Immunol 2008,8:935-947.

50. Zhao X, Deak E, Soderberg K, Linehan M, Spezzano D, Zhu J, et al. Vaginalsubmucosal dendritic cells, but not Langerhans cells, induce protective Th1responses to herpes simplex virus-2. J Exp Med 2003,197:153-162.

51. Mestecky J, Moldoveanu Z, Smith PD, Hel Z, Alexander RC. Mucosal immunology ofthe genital and gastrointestinal tracts and HIV-1 infection. Journal of ReproductiveImmunology 2009,83:196-200.

52. Russell MW, Mestecky J. Humoral immune responses to microbial infections in thegenital tract. Microbes and Infection 2002,4:667-677.

53. Rosenthal KL, Gallichan WS. Challenges for vaccination against sexually-transmitteddiseases: induction and long-term maintenance of mucosal immune responses inthe female genital tract. Seminars in Immunology 1997,9:303-314.

54. MasCasullo V, Fam E, Keller MJ, Herold BC. Role of mucosal immunity in preventinggenital herpes infection. Viral Immunology 2005,18:595-606.

55. Wira CR, Rossoll RM. Antigen-presenting cells in the female reproductive tract:influence of sex hormones on antigen presentation in the vagina. Immunology1995,84:505-508.

56. Wira CR, Rossoll RM, Kaushic C. Antigen-presenting cells in the female reproductivetract: Influence of estradiol on antigen presentation by vaginal cells. Endocrinology2000,141:2877-2885.

57. Wira CR, Fahey JV, Ghosh M, Patel MV, Hickey DK, Ochiel DO. Sex HormoneRegulation of Innate Immunity in the Female Reproductive Tract: The Role ofEpithelial Cells in Balancing Reproductive Potential with Protection against SexuallyTransmitted Pathogens. American Journal of Reproductive Immunology,63:544-565.

58. Kovats S, Carreras E. Regulation of dendritic cell differentiation and function byestrogen receptor ligands. Cellular Immunology 2008,252:81-90.

59. Beagley KW, Gockel CM. Regulation of innate and adaptive immunity by the femalesex hormones oestradiol and progesterone. Fems Immunology and MedicalMicrobiology 2003,38:13-22.

60. Mostad SB, Overbaugh J, DeVange DM, Welch MJ, Chohan B, Mandaliya K, et al.Hormonal contraception, vitamin A deficiency, and other risk factors for sheddingof HIV-1 infected cells from the cervix and vagina. The Lancet 1997,350:922-927.

61. Aflatoonian R, Fazeli A. Toll-like receptors in female reproductive tract and theirmenstrual cycle dependent expression. Journal of Reproductive Immunology2008,77:7-13.

62. Butts CL CK, Warfel J, Belyavskaya E, D'Agnillo F, Sternberg EM. Progesteroneregulation of uterine dendritic cell function in rodents is dependant on the stage ofestrous cycle. In: Mucosal Immunology; 2010. pp. 496-505.



61

63. Black CA, Rohan LC, Cost M, Watkins SC, Draviam R, Alber S, et al. Vaginal mucosaserves as an inductive site for tolerance. Journal of Immunology 2000,165:5077-5083.

64. Franklin H. Bronson CPD, and George D. Snell. Biology of the Laboratory Mouse. In:Biology of the Laboratory Mouse. Edited by Green EL. SECOND EDITION ed. NEWYORK: DOVER PUBLICATIONS, INC.; 1966.

65. Herbison AE. Physiology of the Gonadotropin-Releasing Hormone NeuronalNetwork. In: Knobil and Neill's Physiology of Reproduction (Third Edition). Edited byJimmy DN, Ph.D, Tony MP, et al. St Louis: Academic Press; 2006. pp. 1415-1482.

66. MURR SM, GESCHWIND II, BRADFORD GE. PLASMA LH AND FSH DURINGDIFFERENT OESTROUS CYCLE CONDITIONS IN MICE. J Reprod Fertil 1973,32:221-230.

67. Westwood FR. The Female Rat Reproductive Cycle: A Practical Histological Guide toStaging. Toxicologic Pathology 2008,36:375-384.

68. OECD. Preparation, Reading and Reporting of Vaginal Smears. In: Endocrinedisruption: a guidance document for histologic evaluation of endocrine andreproductive tests

2008; 2008.69. Soderberg KA, Linehan MM, Ruddle NH, Iwasaki A. MAdCAM-1 Expressing Sacral

Lymph Node in the Lymphotoxin {beta}-Deficient Mouse Provides a Site forImmune Generation Following Vaginal Herpes Simplex Virus-2 Infection. J Immunol2004,173:1908-1913.

70. Velardo JT, Dawson AB, Olsen AG, Hisaw FL. Sequence of histological changes in theuterus and vagina of the rat during prologation of pseudopregnancy associatedwith the presence of deciduomata. American Journal of Anatomy 1953,93:273-305.

71. Sonoda Y, Mukaida N, Wang J-b, Shimada-Hiratsuka M, Naito M, Kasahara T, et al.Physiologic Regulation of Postovulatory Neutrophil Migration into Vagina in Miceby a C-X-C Chemokine(s). The Journal of Immunology 1998,160:6159-6165.

72. Jungi TW, Besedovsky HO, Sorkin E, Schardt M. Hormonal control of neutrophilchemotactic activity in the rat vagina. American Journal of Physiology - Regulatory,Integrative and Comparative Physiology 1977,233:R59-R65.

73. Fichorova RN, Trifonova RT, Gilbert RO, Costello CE, Hayes GR, Lucas JJ, et al.Trichomonas vaginalis Lipophosphoglycan Triggers a Selective Upregulation ofCytokines by Human Female Reproductive Tract Epithelial Cells. Infect. Immun.2006,74:5773-5779.

74. Wood GA, Fata JE, Watson KLM, Khokha R. Circulating hormones and estrous stagepredict cellular and stromal remodeling in murine uterus. Reproduction2007,133:1035-1044.

75. Parr EL, Parr MB. Immune responses and protection against vaginal infection afternasal or vaginal immunization with attenuated herpes simplex virus type-2.Immunology 1999,98:639-645.

76. Kaushic C, Ashkar AA, Reid LA, Rosenthal KL. Progesterone Increases Susceptibilityand Decreases Immune Responses to Genital Herpes Infection. Journal of Virology2003,77:4558-4565.

77. Kaushic C, Roth KL, Anipindi V, Xiu F. Increased prevalence of sexually transmittedviral infections in women: the role of female sex hormones in regulatingsusceptibility and immune responses. Journal of Reproductive Immunology2011,88:204-209.



62

78. Kaushic C. The role of the local microenvironment in regulating susceptibility andimmune responses to sexually transmitted viruses in the female genital tract.Journal of Reproductive Immunology 2009,83:168-172.

79. Winterhager E, Kühnel W. Diffusion barriers in the vaginal epithelium during theestrous cycle in guinea pigs. Cell and Tissue Research 1985,241:325-331.

80. Koiter TR, Hazenberg MP, Van der Schoot P. Regulation of the bacterial microfloraof the vagina in cyclic female rats. Journal of Experimental Zoology 1977,202:121-127.

81. Iijima N, Thompson JM, Iwasaki A. Dendritic cells and macrophages in thegenitourinary tract. Mucosal Immunol 2008,1:451-459.

82. Kaushic C, Frauendorf E, Rossoll RM, Richardson JM, Wira CR. Influence of theestrous cycle on the presence and distribution of immune cells in the ratreproductive tract. Am J Reprod Immunol 1998,39:209-216.

83. Wieser F, Hosmann J, Tschugguel W, Czerwenka K, Sedivy R, Huber JC.Progesterone increases the number of Langerhans cells in human vaginalepithelium. Fertility and Sterility 2001,75:1234-1235.

84. Johansson EL, Rudin A, Wassen L, Holmgren J. Distribution of lymphocytes andadhesion molecules in human cervix and vagina. Immunology 1999,96:272-277.

85. Schulz O, Jaensson E, Persson EK, Liu X, Worbs T, Agace WW, et al. IntestinalCD103+, but not CX3CR1+, antigen sampling cells migrate in lymph and serveclassical dendritic cell functions. The Journal of Experimental Medicine2009,206:3101-3114.

86. Rescigno M, Urbano M, Valzasina B, Francolini M, Rotta G, Bonasio R, et al.Dendritic cells express tight junction proteins and penetrate gut epithelialmonolayers to sample bacteria. Nat Immunol 2001,2:361-367.

87. Niedergang F, Kweon M-N. New trends in antigen uptake in the gut mucosa. Trendsin Microbiology 2005,13:485-490.

88. LeBlanc DM, Barousse MM, Fidel PL, Jr. Role for Dendritic Cells inImmunoregulation during Experimental Vaginal Candidiasis. Infect. Immun.2006,74:3213-3221.

89. Nestle FO, Di Meglio P, Qin J-Z, Nickoloff BJ. Skin immune sentinels in health anddisease. Nat Rev Immunol 2009,9:679-691.

90. Mosser DM, Edwards JP. Exploring the full spectrum of macrophage activation. NatRev Immunol 2008,8:958-969.

91. Hume DA. The mononuclear phagocyte system. Current Opinion in Immunology2006,18:49-53.

92. Shen R, Richter HE, Clements RH, Novak L, Huff K, Bimczok D, et al. Macrophages inVaginal but Not Intestinal Mucosa Are Monocyte-Like and Permissive to HumanImmunodeficiency Virus Type 1 Infection. Journal of Virology 2009,83:3258-3267.

93. Horton RE, Kaefer N, Songok E, Guijon FB, Kettaf N, Boucher G, et al. A ComparativeAnalysis of Gene Expression Patterns and Cell Phenotypes between Cervical andPeripheral Blood Mononuclear Cells. PLoS One 2009,4:e8293.

94. Hume DA. Macrophages as APC and the Dendritic Cell Myth. The Journal ofImmunology 2008,181:5829-5835.

95. Pudney J, Quayle AJ, Anderson DJ. Immunological Microenvironments in theHuman Vagina and Cervix: Mediators of Cellular Immunity Are Concentrated in theCervical Transformation Zone. Biology of Reproduction 2005,73:1253-1263.

96. Egan C, Sukhumavasi W, Bierly A, Denkers E. Understanding the multiple functionsof Gr-1+ cell subpopulations during microbial infection. Immunologic Research2008,40:35-48.



63

97. Bennouna S, Bliss SK, Curiel TJ, Denkers EY. Cross-Talk in the Innate ImmuneSystem: Neutrophils Instruct Recruitment and Activation of Dendritic Cells duringMicrobial Infection. J Immunol 2003,171:6052-6058.

98. Ludwig IS, Geijtenbeek TBH, van Kooyk Y. Two way communication betweenneutrophils and dendritic cells. Current Opinion in Pharmacology 2006,6:408-413.

99. Megiovanni AM, Sanchez F, Robledo-Sarmiento M, Morel C, Gluckman JC, BoudalyS. Polymorphonuclear neutrophils deliver activation signals and antigenicmolecules to dendritic cells: a new link between leukocytes upstream of Tlymphocytes. J Leukoc Biol 2006,79:977-988.

100. Morel C, Badell E, Abadie V, Robledo M, Setterblad N, Gluckman JC, et al.Mycobacterium bovis BCG-infected neutrophils and dendritic cells cooperate toinduce specific T cell responses in humans and mice. European Journal ofImmunology 2008,38:437-447.

101. van Gisbergen KPJM, Sanchez-Hernandez M, Geijtenbeek TBH, van Kooyk Y.Neutrophils mediate immune modulation of dendritic cells through glycosylation-dependent interactions between Mac-1 and DC-SIGN. J. Exp. Med. 2005,201:1281-1292.

102. Gallichan WS, Rosenthal KL. Long-lived cytotoxic T lymphocyte memory in mucosaltissues after mucosal but not systemic immunization. The Journal of ExperimentalMedicine 1996,184:1879-1890.

103. Kaufman DR, Barouch DH. Translational Mini-Review Series on Vaccines for HIV: Tlymphocyte trafficking and vaccine-elicited mucosal immunity. Clinical &Experimental Immunology 2009,157:165-173.

104. Mebius RE. Organogenesis of lymphoid tissues. Nat Rev Immunol 2003,3:292-303.105. Randall TD, Carragher DM, Rangel-Moreno J. Development of Secondary Lymphoid

Organs. Annual Review of Immunology 2008,26:627-650.106. Müller G, Lipp M. Concerted action of the chemokine and lymphotoxin system in

secondary lymphoid-organ development. Current Opinion in Immunology2003,15:217-224.

107. Rennert PD, Browning JL, Hochman PS. Selective disruption of lymphotoxin ligandsreveals a novel set of mucosal lymph nodes and unique effects on lymph nodecellular organization. International Immunology 1997,9:1627-1639.

108. Alimzhanov MB, Kuprash DV, Kosco-Vilbois MH, Luz A, Turetskaya RL, TarakhovskyA, et al. Abnormal development of secondary lymphoid tissues in lymphotoxin β-deficient mice. Proceedings of the National Academy of Sciences 1997,94:9302-9307.

109. Chtanova T, Han S-J, Schaeffer M, van Dooren GG, Herzmark P, Striepen B, et al.Dynamics of T Cell, Antigen-Presenting Cell, and Pathogen Interactions duringRecall Responses in the Lymph Node. Immunity 2009,31:342-355.

110. Jaensson E, Uronen-Hansson H, Pabst O, Eksteen B, Tian J, Coombes JL, et al. Smallintestinal CD103+ dendritic cells display unique functional properties that areconserved between mice and humans. The Journal of Experimental Medicine2008,205:2139-2149.

111. Kim TS, Hufford MM, Sun J, Fu Y-X, Braciale TJ. Antigen persistence and the controlof local T cell memory by migrant respiratory dendritic cells after acute virusinfection. The Journal of Experimental Medicine 2010,207:1161-1172.

112. Bode U, Lorchner M, Ahrendt M, Blessenohl M, Kalies K, Claus A, et al. Dendriticcell subsets in lymph nodes are characterized by the specific draining area andinfluence the phenotype and fate of primed T cells. Immunology 2008,123:480-490.



64

113. Guilliams M, Crozat K, Henri S, Tamoutounour S, Grenot P, Devilard E, et al. Skin-draining lymph nodes contain dermis-derived CD103- dendritic cells thatconstitutively produce retinoic acid and induce Foxp3+ regulatory T cells. Blood2010,115:1958-1968.

114. Macpherson AJ, Smith K. Mesenteric lymph nodes at the center of immuneanatomy. The Journal of Experimental Medicine 2006,203:497-500.

115. von Andrian UH, Mempel TR. Homing and cellular traffic in lymph nodes. Nat RevImmunol 2003,3:867-878.

116. Weninger W, Crowley MA, Manjunath N, von Andrian UH. Migratory Properties ofNaive, Effector, and Memory Cd8+ T Cells. The Journal of Experimental Medicine2001,194:953-966.

117. Buettner M, Pabst R, Bode U. Stromal cell heterogeneity in lymphoid organs.Trends in Immunology,31:80-86.

118. Sigmundsdottir H, Butcher EC. Environmental cues, dendritic cells and theprogramming of tissue-selective lymphocyte trafficking. Nat Immunol 2008,9:981-987.

119. Turley SJ, Fletcher AL, Elpek KG. The stromal and haematopoietic antigen-presenting cells that reside in secondary lymphoid organs. Nat Rev Immunol2010,10:813-825.

120. Van den Broeck W, Derore A, Simoens P. Anatomy and nomenclature of murinelymph nodes: Descriptive study and nomenclatory standardization in BALB/cAnNCrlmice. Journal of Immunological Methods 2006,312:12-19.

121. Salomon B, Cohen JL, Masurier C, Klatzmann D. Three Populations of Mouse LymphNode Dendritic Cells with Different Origins and Dynamics. The Journal ofImmunology 1998,160:708-717.

122. Vermaelen KY, Carro-Muino I, Lambrecht BN, Pauwels RA. Specific MigratoryDendritic Cells Rapidly Transport Antigen from the Airways to the Thoracic LymphNodes. The Journal of Experimental Medicine 2001,193:51-60.

123. Iwasaki A, Kelsall BL. Unique functions of CD11b+, CD8 alpha+, and double-negativePeyer's patch dendritic cells. J Immunol 2001,166:4884-4890.

124. Naik SH. Demystifying the development of dendritic cell subtypes, a little. ImmunolCell Biol 2008,86:439-452.

125. Shortman K, Liu Y-J. Mouse and human dendritic cell subtypes. Nat Rev Immunol2002,2:151-161.

126. Coombes JL, Powrie F. Dendritic cells in intestinal immune regulation. Nat RevImmunol 2008,8:435-446.

127. Shortman K, Naik SH. Steady-state and inflammatory dendritic-cell development.Nat Rev Immunol 2007,7:19-30.

128. A Ballesteros-Tato BL, FE Lund and TD Randall. Temporal changes in dendritic cellsubsets, cross-priming and costimulation via CD70 control CD8+ T cell responses toinfluenza. In: Nature Immunology; 2010. pp. 216-224.

129. Laffont S, Siddiqui KRR, Powrie F. Intestinal inflammation abrogates the tolerogenicproperties of MLN CD103+ dendritic cells. European Journal of Immunology2010,40:1877-1883.

130. King IL, Kroenke MA, Segal BM. GM-CSF–dependent, CD103+ dermal dendritic cellsplay a critical role in Th effector cell differentiation after subcutaneousimmunization. The Journal of Experimental Medicine 2010,207:953-961.

131. Annacker O, Coombes JL, Malmstrom V, Uhlig HH, Bourne T, Johansson-Lindbom B,et al. Essential role for CD103 in the T cell-mediated regulation of experimentalcolitis. J Exp Med 2005,202:1051-1061.



65

132. Pulendran B, Tang H, Manicassamy S. Programming dendritic cells to induce T(H)2and tolerogenic responses. Nature Immunology 2010,11:647-655.

133. Manicassamy S, Pulendran B. Retinoic acid-dependent regulation of immuneresponses by dendritic cells and macrophages. Seminars in Immunology2009,21:22-27.

134. Mora JR, Iwata M, von Andrian UH. Vitamin effects on the immune system:vitamins A and D take centre stage. Nat Rev Immunol 2008,8:685-698.

135. Smith JM, Wira CR, Fanger MW, Shen L. Human fallopian tube neutrophils - Adistinct phenotype from blood neutrophils. American Journal of ReproductiveImmunology 2006,56:218-229.

136. Sasaki S, Nagata K, Kobayashi Y. Regulation of the estrous cycle by neutrophilinfiltration into the vagina. Biochemical and Biophysical Research Communications2009,382:35-40.

137. De Santo C, Arscott R, Booth S, Karydis I, Jones M, Asher R, et al. Invariant NKT cellsmodulate the suppressive activity of IL-10-secreting neutrophils differentiated withserum amyloid A. Nat Immunol 2010,11:1039-1046.

138. Johansson-Lindbom B, Svensson M, Pabst O, Palmqvist C, Marquez G, Förster R, etal. Functional specialization of gut CD103+ dendritic cells in the regulation oftissue-selective T cell homing. The Journal of Experimental Medicine2005,202:1063-1073.

139. Patton DL, Thwin SS, Meier A, Hooton TM, Stapleton AE, Eschenbach DA. Epithelialcell layer thickness and immune cell populations in the normal human vagina atdifferent stages of the menstrual cycle. American Journal of Obstetrics andGynecology 2000,183:967-973.

characterisation of antigen-presenting cells in the murine...

Documents